id
large_stringlengths 9
16
| submitter
large_stringlengths 3
64
⌀ | authors
large_stringlengths 5
20.2k
| title
large_stringlengths 5
245
| comments
large_stringlengths 1
482
⌀ | journal-ref
large_stringlengths 4
382
⌀ | doi
large_stringlengths 9
151
⌀ | report-no
large_stringlengths 2
509
⌀ | categories
large_stringlengths 5
78
| license
large_stringclasses 9
values | abstract
large_stringlengths 83
3.55k
| versions
large listlengths 1
21
| update_date
timestamp[ms]date 2007-05-23 00:00:00
2025-05-30 00:00:00
| authors_parsed
large listlengths 1
1.35k
| prompt
large_stringlengths 166
3.64k
| label
stringclasses 2
values | prob
float64 0.5
0.73
|
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
2111.11343 | Jesse Cresswell | Shivam Kalra, Junfeng Wen, Jesse C. Cresswell, Maksims Volkovs, Hamid
R. Tizhoosh | Decentralized Federated Learning through Proxy Model Sharing | null | Nature Communications 14, 2899 (2023) | 10.1038/s41467-023-38569-4 | null | cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Institutions in highly regulated domains such as finance and healthcare often
have restrictive rules around data sharing. Federated learning is a distributed
learning framework that enables multi-institutional collaborations on
decentralized data with improved protection for each collaborator's data
privacy. In this paper, we propose a communication-efficient scheme for
decentralized federated learning called ProxyFL, or proxy-based federated
learning. Each participant in ProxyFL maintains two models, a private model,
and a publicly shared proxy model designed to protect the participant's
privacy. Proxy models allow efficient information exchange among participants
without the need of a centralized server. The proposed method eliminates a
significant limitation of canonical federated learning by allowing model
heterogeneity; each participant can have a private model with any architecture.
Furthermore, our protocol for communication by proxy leads to stronger privacy
guarantees using differential privacy analysis. Experiments on popular image
datasets, and a cancer diagnostic problem using high-quality gigapixel
histology whole slide images, show that ProxyFL can outperform existing
alternatives with much less communication overhead and stronger privacy.
| [
{
"version": "v1",
"created": "Mon, 22 Nov 2021 16:47:39 GMT"
},
{
"version": "v2",
"created": "Tue, 23 May 2023 02:18:27 GMT"
}
] | 2023-05-24T00:00:00 | [
[
"Kalra",
"Shivam",
""
],
[
"Wen",
"Junfeng",
""
],
[
"Cresswell",
"Jesse C.",
""
],
[
"Volkovs",
"Maksims",
""
],
[
"Tizhoosh",
"Hamid R.",
""
]
] | TITLE: Decentralized Federated Learning through Proxy Model Sharing
ABSTRACT: Institutions in highly regulated domains such as finance and healthcare often
have restrictive rules around data sharing. Federated learning is a distributed
learning framework that enables multi-institutional collaborations on
decentralized data with improved protection for each collaborator's data
privacy. In this paper, we propose a communication-efficient scheme for
decentralized federated learning called ProxyFL, or proxy-based federated
learning. Each participant in ProxyFL maintains two models, a private model,
and a publicly shared proxy model designed to protect the participant's
privacy. Proxy models allow efficient information exchange among participants
without the need of a centralized server. The proposed method eliminates a
significant limitation of canonical federated learning by allowing model
heterogeneity; each participant can have a private model with any architecture.
Furthermore, our protocol for communication by proxy leads to stronger privacy
guarantees using differential privacy analysis. Experiments on popular image
datasets, and a cancer diagnostic problem using high-quality gigapixel
histology whole slide images, show that ProxyFL can outperform existing
alternatives with much less communication overhead and stronger privacy.
| no_new_dataset | 0.711606 |
2111.15454 | Siyuan Li | Siyuan Li, Zicheng Liu, Zedong Wang, Di Wu, Zihan Liu, Stan Z. Li | Boosting Discriminative Visual Representation Learning with
Scenario-Agnostic Mixup | Preprint version v3 with 9 pages main body and 8 pages appendix. The
source code is available at \url{https://github.com/Westlake-AI/openmixup} | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Mixup is a well-known data-dependent augmentation technique for DNNs,
consisting of two sub-tasks: mixup generation and classification. However, the
recent dominant online training method confines mixup to supervised learning
(SL), and the objective of the generation sub-task is limited to selected
sample pairs instead of the whole data manifold, which might cause trivial
solutions. To overcome such limitations, we comprehensively study the objective
of mixup generation and propose \textbf{S}cenario-\textbf{A}gnostic
\textbf{Mix}up (SAMix) for both SL and Self-supervised Learning (SSL)
scenarios. Specifically, we hypothesize and verify the objective function of
mixup generation as optimizing local smoothness between two mixed classes
subject to global discrimination from other classes. Accordingly, we propose
$\eta$-balanced mixup loss for complementary learning of the two
sub-objectives. Meanwhile, a label-free generation sub-network is designed,
which effectively provides non-trivial mixup samples and improves transferable
abilities. Moreover, to reduce the computational cost of online training, we
further introduce a pre-trained version, SAMix$^\mathcal{P}$, achieving more
favorable efficiency and generalizability. Extensive experiments on nine SL and
SSL benchmarks demonstrate the consistent superiority and versatility of SAMix
compared with existing methods.
| [
{
"version": "v1",
"created": "Tue, 30 Nov 2021 14:49:59 GMT"
},
{
"version": "v2",
"created": "Thu, 29 Sep 2022 20:19:27 GMT"
},
{
"version": "v3",
"created": "Tue, 23 May 2023 17:51:28 GMT"
}
] | 2023-05-24T00:00:00 | [
[
"Li",
"Siyuan",
""
],
[
"Liu",
"Zicheng",
""
],
[
"Wang",
"Zedong",
""
],
[
"Wu",
"Di",
""
],
[
"Liu",
"Zihan",
""
],
[
"Li",
"Stan Z.",
""
]
] | TITLE: Boosting Discriminative Visual Representation Learning with
Scenario-Agnostic Mixup
ABSTRACT: Mixup is a well-known data-dependent augmentation technique for DNNs,
consisting of two sub-tasks: mixup generation and classification. However, the
recent dominant online training method confines mixup to supervised learning
(SL), and the objective of the generation sub-task is limited to selected
sample pairs instead of the whole data manifold, which might cause trivial
solutions. To overcome such limitations, we comprehensively study the objective
of mixup generation and propose \textbf{S}cenario-\textbf{A}gnostic
\textbf{Mix}up (SAMix) for both SL and Self-supervised Learning (SSL)
scenarios. Specifically, we hypothesize and verify the objective function of
mixup generation as optimizing local smoothness between two mixed classes
subject to global discrimination from other classes. Accordingly, we propose
$\eta$-balanced mixup loss for complementary learning of the two
sub-objectives. Meanwhile, a label-free generation sub-network is designed,
which effectively provides non-trivial mixup samples and improves transferable
abilities. Moreover, to reduce the computational cost of online training, we
further introduce a pre-trained version, SAMix$^\mathcal{P}$, achieving more
favorable efficiency and generalizability. Extensive experiments on nine SL and
SSL benchmarks demonstrate the consistent superiority and versatility of SAMix
compared with existing methods.
| no_new_dataset | 0.708799 |
2201.02949 | Enes Altinisik | Enes Altinisik, Husrev Taha Sencar, Diram Tabaa | Video Source Characterization Using Encoding and Encapsulation
Characteristics | null | null | 10.1109/TIFS.2022.3204210 | null | cs.CR cs.MM | http://creativecommons.org/licenses/by-nc-nd/4.0/ | We introduce a new method for camera-model identification. Our approach
combines two independent aspects of video file generation corresponding to
video coding and media data encapsulation. To this end, a joint representation
of the overall file metadata is developed and used in conjunction with a
two-level hierarchical classification method. At the first level, our method
groups videos into metaclasses considering several abstractions that represent
high-level structural properties of file metadata. This is followed by a more
nuanced classification of classes that comprise each metaclass. The method is
evaluated on more than 20K videos obtained by combining four public video
datasets. Tests show that a balanced accuracy of 91% is achieved in correctly
identifying the class of a video among 119 video classes. This corresponds to
an improvement of 6.5% over the conventional approach based on video file
encapsulation characteristics. Furthermore, we investigate a setting relevant
to forensic file recovery operations where file metadata cannot be located or
are missing but video data is partially available. By estimating a partial list
of encoding parameters from coded video data, we demonstrate that an
identification accuracy of 57% can be achieved in camera-model identification
in the absence of any other file metadata.
| [
{
"version": "v1",
"created": "Sun, 9 Jan 2022 07:35:07 GMT"
},
{
"version": "v2",
"created": "Fri, 4 Feb 2022 07:41:23 GMT"
},
{
"version": "v3",
"created": "Wed, 24 Aug 2022 13:16:40 GMT"
},
{
"version": "v4",
"created": "Sun, 28 Aug 2022 11:44:50 GMT"
}
] | 2023-05-24T00:00:00 | [
[
"Altinisik",
"Enes",
""
],
[
"Sencar",
"Husrev Taha",
""
],
[
"Tabaa",
"Diram",
""
]
] | TITLE: Video Source Characterization Using Encoding and Encapsulation
Characteristics
ABSTRACT: We introduce a new method for camera-model identification. Our approach
combines two independent aspects of video file generation corresponding to
video coding and media data encapsulation. To this end, a joint representation
of the overall file metadata is developed and used in conjunction with a
two-level hierarchical classification method. At the first level, our method
groups videos into metaclasses considering several abstractions that represent
high-level structural properties of file metadata. This is followed by a more
nuanced classification of classes that comprise each metaclass. The method is
evaluated on more than 20K videos obtained by combining four public video
datasets. Tests show that a balanced accuracy of 91% is achieved in correctly
identifying the class of a video among 119 video classes. This corresponds to
an improvement of 6.5% over the conventional approach based on video file
encapsulation characteristics. Furthermore, we investigate a setting relevant
to forensic file recovery operations where file metadata cannot be located or
are missing but video data is partially available. By estimating a partial list
of encoding parameters from coded video data, we demonstrate that an
identification accuracy of 57% can be achieved in camera-model identification
in the absence of any other file metadata.
| no_new_dataset | 0.710201 |
2201.13405 | Evgeniia Razumovskaia | Olga Majewska, Evgeniia Razumovskaia, Edoardo Maria Ponti, Ivan
Vuli\'c, Anna Korhonen | Cross-Lingual Dialogue Dataset Creation via Outline-Based Generation | null | null | 10.1162/tacl_a_00539 | null | cs.CL | http://creativecommons.org/licenses/by/4.0/ | Multilingual task-oriented dialogue (ToD) facilitates access to services and
information for many (communities of) speakers. Nevertheless, the potential of
this technology is not fully realised, as current datasets for multilingual ToD
- both for modular and end-to-end modelling - suffer from severe limitations.
1) When created from scratch, they are usually small in scale and fail to cover
many possible dialogue flows. 2) Translation-based ToD datasets might lack
naturalness and cultural specificity in the target language. In this work, to
tackle these limitations we propose a novel outline-based annotation process
for multilingual ToD datasets, where domain-specific abstract schemata of
dialogue are mapped into natural language outlines. These in turn guide the
target language annotators in writing a dialogue by providing instructions
about each turn's intents and slots. Through this process we annotate a new
large-scale dataset for training and evaluation of multilingual and
cross-lingual ToD systems. Our Cross-lingual Outline-based Dialogue dataset
(termed COD) enables natural language understanding, dialogue state tracking,
and end-to-end dialogue modelling and evaluation in 4 diverse languages:
Arabic, Indonesian, Russian, and Kiswahili. Qualitative and quantitative
analyses of COD versus an equivalent translation-based dataset demonstrate
improvements in data quality, unlocked by the outline-based approach. Finally,
we benchmark a series of state-of-the-art systems for cross-lingual ToD,
setting reference scores for future work and demonstrating that COD prevents
over-inflated performance, typically met with prior translation-based ToD
datasets.
| [
{
"version": "v1",
"created": "Mon, 31 Jan 2022 18:11:21 GMT"
}
] | 2023-05-24T00:00:00 | [
[
"Majewska",
"Olga",
""
],
[
"Razumovskaia",
"Evgeniia",
""
],
[
"Ponti",
"Edoardo Maria",
""
],
[
"Vulić",
"Ivan",
""
],
[
"Korhonen",
"Anna",
""
]
] | TITLE: Cross-Lingual Dialogue Dataset Creation via Outline-Based Generation
ABSTRACT: Multilingual task-oriented dialogue (ToD) facilitates access to services and
information for many (communities of) speakers. Nevertheless, the potential of
this technology is not fully realised, as current datasets for multilingual ToD
- both for modular and end-to-end modelling - suffer from severe limitations.
1) When created from scratch, they are usually small in scale and fail to cover
many possible dialogue flows. 2) Translation-based ToD datasets might lack
naturalness and cultural specificity in the target language. In this work, to
tackle these limitations we propose a novel outline-based annotation process
for multilingual ToD datasets, where domain-specific abstract schemata of
dialogue are mapped into natural language outlines. These in turn guide the
target language annotators in writing a dialogue by providing instructions
about each turn's intents and slots. Through this process we annotate a new
large-scale dataset for training and evaluation of multilingual and
cross-lingual ToD systems. Our Cross-lingual Outline-based Dialogue dataset
(termed COD) enables natural language understanding, dialogue state tracking,
and end-to-end dialogue modelling and evaluation in 4 diverse languages:
Arabic, Indonesian, Russian, and Kiswahili. Qualitative and quantitative
analyses of COD versus an equivalent translation-based dataset demonstrate
improvements in data quality, unlocked by the outline-based approach. Finally,
we benchmark a series of state-of-the-art systems for cross-lingual ToD,
setting reference scores for future work and demonstrating that COD prevents
over-inflated performance, typically met with prior translation-based ToD
datasets.
| new_dataset | 0.712238 |
2202.06658 | Bin Liu | Hao Guo, Jiyong Jin, Bin Liu | PFGE: Parsimonious Fast Geometric Ensembling of DNNs | This paper has been accepted by 19th Inter. Conf. on Intelligent
Computing (ICIC 2023) | null | null | null | cs.LG cs.AI | http://creativecommons.org/licenses/by-nc-nd/4.0/ | Ensemble methods are commonly used to enhance the generalization performance
of machine learning models. However, they present a challenge in deep learning
systems due to the high computational overhead required to train an ensemble of
deep neural networks (DNNs). Recent advancements such as fast geometric
ensembling (FGE) and snapshot ensembles have addressed this issue by training
model ensembles in the same time as a single model. Nonetheless, these
techniques still require additional memory for test-time inference compared to
single-model-based methods. In this paper, we propose a new method called
parsimonious FGE (PFGE), which employs a lightweight ensemble of
higher-performing DNNs generated through successive stochastic weight averaging
procedures. Our experimental results on CIFAR-{10,100} and ImageNet datasets
across various modern DNN architectures demonstrate that PFGE achieves 5x
memory efficiency compared to previous methods, without compromising on
generalization performance. For those interested, our code is available at
https://github.com/ZJLAB-AMMI/PFGE.
| [
{
"version": "v1",
"created": "Mon, 14 Feb 2022 12:27:46 GMT"
},
{
"version": "v2",
"created": "Tue, 22 Feb 2022 11:44:07 GMT"
},
{
"version": "v3",
"created": "Wed, 23 Feb 2022 06:35:33 GMT"
},
{
"version": "v4",
"created": "Sun, 24 Apr 2022 08:12:07 GMT"
},
{
"version": "v5",
"created": "Wed, 18 May 2022 10:11:23 GMT"
},
{
"version": "v6",
"created": "Tue, 11 Oct 2022 02:35:17 GMT"
},
{
"version": "v7",
"created": "Sat, 18 Feb 2023 08:47:50 GMT"
},
{
"version": "v8",
"created": "Tue, 23 May 2023 08:53:32 GMT"
}
] | 2023-05-24T00:00:00 | [
[
"Guo",
"Hao",
""
],
[
"Jin",
"Jiyong",
""
],
[
"Liu",
"Bin",
""
]
] | TITLE: PFGE: Parsimonious Fast Geometric Ensembling of DNNs
ABSTRACT: Ensemble methods are commonly used to enhance the generalization performance
of machine learning models. However, they present a challenge in deep learning
systems due to the high computational overhead required to train an ensemble of
deep neural networks (DNNs). Recent advancements such as fast geometric
ensembling (FGE) and snapshot ensembles have addressed this issue by training
model ensembles in the same time as a single model. Nonetheless, these
techniques still require additional memory for test-time inference compared to
single-model-based methods. In this paper, we propose a new method called
parsimonious FGE (PFGE), which employs a lightweight ensemble of
higher-performing DNNs generated through successive stochastic weight averaging
procedures. Our experimental results on CIFAR-{10,100} and ImageNet datasets
across various modern DNN architectures demonstrate that PFGE achieves 5x
memory efficiency compared to previous methods, without compromising on
generalization performance. For those interested, our code is available at
https://github.com/ZJLAB-AMMI/PFGE.
| no_new_dataset | 0.709372 |
2204.02450 | Xuanang Xu | Xuanang Xu, Hannah H. Deng, Tianyi Chen, Tianshu Kuang, Joshua C.
Barber, Daeseung Kim, Jaime Gateno, James J. Xia, Pingkun Yan | Federated Cross Learning for Medical Image Segmentation | v1: 10 pages, 4 figures; v2: 12 pages, 5 figures, accepted by Medical
Imaging with Deep Learning (MIDL) 2023 conference, camera-ready version
available at https://openreview.net/forum?id=DrZbwobH_zo , source code
available at https://github.com/DIAL-RPI/FedCross | null | null | null | eess.IV cs.CV | http://creativecommons.org/licenses/by/4.0/ | Federated learning (FL) can collaboratively train deep learning models using
isolated patient data owned by different hospitals for various clinical
applications, including medical image segmentation. However, a major problem of
FL is its performance degradation when dealing with data that are not
independently and identically distributed (non-iid), which is often the case in
medical images. In this paper, we first conduct a theoretical analysis on the
FL algorithm to reveal the problem of model aggregation during training on
non-iid data. With the insights gained through the analysis, we propose a
simple yet effective method, federated cross learning (FedCross), to tackle
this challenging problem. Unlike the conventional FL methods that combine
multiple individually trained local models on a server node, our FedCross
sequentially trains the global model across different clients in a round-robin
manner, and thus the entire training procedure does not involve any model
aggregation steps. To further improve its performance to be comparable with the
centralized learning method, we combine the FedCross with an ensemble learning
mechanism to compose a federated cross ensemble learning (FedCrossEns) method.
Finally, we conduct extensive experiments using a set of public datasets. The
experimental results show that the proposed FedCross training strategy
outperforms the mainstream FL methods on non-iid data. In addition to improving
the segmentation performance, our FedCrossEns can further provide a
quantitative estimation of the model uncertainty, demonstrating the
effectiveness and clinical significance of our designs. Source code is publicly
available at https://github.com/DIAL-RPI/FedCross.
| [
{
"version": "v1",
"created": "Tue, 5 Apr 2022 18:55:02 GMT"
},
{
"version": "v2",
"created": "Mon, 22 May 2023 21:05:06 GMT"
}
] | 2023-05-24T00:00:00 | [
[
"Xu",
"Xuanang",
""
],
[
"Deng",
"Hannah H.",
""
],
[
"Chen",
"Tianyi",
""
],
[
"Kuang",
"Tianshu",
""
],
[
"Barber",
"Joshua C.",
""
],
[
"Kim",
"Daeseung",
""
],
[
"Gateno",
"Jaime",
""
],
[
"Xia",
"James J.",
""
],
[
"Yan",
"Pingkun",
""
]
] | TITLE: Federated Cross Learning for Medical Image Segmentation
ABSTRACT: Federated learning (FL) can collaboratively train deep learning models using
isolated patient data owned by different hospitals for various clinical
applications, including medical image segmentation. However, a major problem of
FL is its performance degradation when dealing with data that are not
independently and identically distributed (non-iid), which is often the case in
medical images. In this paper, we first conduct a theoretical analysis on the
FL algorithm to reveal the problem of model aggregation during training on
non-iid data. With the insights gained through the analysis, we propose a
simple yet effective method, federated cross learning (FedCross), to tackle
this challenging problem. Unlike the conventional FL methods that combine
multiple individually trained local models on a server node, our FedCross
sequentially trains the global model across different clients in a round-robin
manner, and thus the entire training procedure does not involve any model
aggregation steps. To further improve its performance to be comparable with the
centralized learning method, we combine the FedCross with an ensemble learning
mechanism to compose a federated cross ensemble learning (FedCrossEns) method.
Finally, we conduct extensive experiments using a set of public datasets. The
experimental results show that the proposed FedCross training strategy
outperforms the mainstream FL methods on non-iid data. In addition to improving
the segmentation performance, our FedCrossEns can further provide a
quantitative estimation of the model uncertainty, demonstrating the
effectiveness and clinical significance of our designs. Source code is publicly
available at https://github.com/DIAL-RPI/FedCross.
| no_new_dataset | 0.710427 |
2204.12907 | Ebrahim Kolahdouz | Ebrahim M. Kolahdouz, David R. Wells, Simone Rossi, Kenneth I. Aycock,
Brent A. Craven and Boyce E. Griffith | A sharp interface Lagrangian-Eulerian method for flexible-body
fluid-structure interaction | null | null | 10.1016/j.jcp.2023.112174 | null | math.NA cs.NA math-ph math.MP physics.flu-dyn | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | This paper introduces a sharp-interface approach to simulating
fluid-structure interaction involving flexible bodies described by general
nonlinear material models and across a broad range of mass density ratios. This
new flexible-body immersed Lagrangian-Eulerian (ILE) approach incorporates the
geometrical and domain solution flexibility of the immersed boundary (IB)
method with an accuracy comparable to body-fitted approaches that sharply
resolve flows and stresses up to the fluid-structure interface. Unlike many IB
methods, our ILE formulation uses distinct momentum equations for the fluid and
solid subregions with a Dirichlet-Neumann coupling strategy that connects fluid
and solid subproblems through simple interface conditions. We use a penalty
method involving two representations of the fluid-structure interface. These
two representations are connected by approximate Lagrange multiplier forces
that impose kinematic interface conditions. This approach also enables the use
of multi-rate time stepping, which allows us to take different time step sizes
for the fluid and structure subproblems. Our fluid solver relies on an immersed
interface method for discrete surfaces to impose stress jump conditions along
complex interfaces. The dynamics of the volumetric structural mesh are
determined using a standard finite element approach to large-deformation
nonlinear elasticity via a nearly incompressible solid mechanics formulation.
This formulation also readily accommodates compressible structures for cases in
which part of the solid boundary does not contact the incompressible fluid.
Comparisons are made with computational and experimental benchmarks. We also
demonstrate the capabilities of this methodology by applying it to model the
transport and capture of a cylindrical blood clot in an inferior vena cava
filter.
| [
{
"version": "v1",
"created": "Wed, 27 Apr 2022 13:18:30 GMT"
},
{
"version": "v2",
"created": "Wed, 22 Jun 2022 20:56:23 GMT"
},
{
"version": "v3",
"created": "Sun, 29 Jan 2023 18:14:54 GMT"
}
] | 2023-05-24T00:00:00 | [
[
"Kolahdouz",
"Ebrahim M.",
""
],
[
"Wells",
"David R.",
""
],
[
"Rossi",
"Simone",
""
],
[
"Aycock",
"Kenneth I.",
""
],
[
"Craven",
"Brent A.",
""
],
[
"Griffith",
"Boyce E.",
""
]
] | TITLE: A sharp interface Lagrangian-Eulerian method for flexible-body
fluid-structure interaction
ABSTRACT: This paper introduces a sharp-interface approach to simulating
fluid-structure interaction involving flexible bodies described by general
nonlinear material models and across a broad range of mass density ratios. This
new flexible-body immersed Lagrangian-Eulerian (ILE) approach incorporates the
geometrical and domain solution flexibility of the immersed boundary (IB)
method with an accuracy comparable to body-fitted approaches that sharply
resolve flows and stresses up to the fluid-structure interface. Unlike many IB
methods, our ILE formulation uses distinct momentum equations for the fluid and
solid subregions with a Dirichlet-Neumann coupling strategy that connects fluid
and solid subproblems through simple interface conditions. We use a penalty
method involving two representations of the fluid-structure interface. These
two representations are connected by approximate Lagrange multiplier forces
that impose kinematic interface conditions. This approach also enables the use
of multi-rate time stepping, which allows us to take different time step sizes
for the fluid and structure subproblems. Our fluid solver relies on an immersed
interface method for discrete surfaces to impose stress jump conditions along
complex interfaces. The dynamics of the volumetric structural mesh are
determined using a standard finite element approach to large-deformation
nonlinear elasticity via a nearly incompressible solid mechanics formulation.
This formulation also readily accommodates compressible structures for cases in
which part of the solid boundary does not contact the incompressible fluid.
Comparisons are made with computational and experimental benchmarks. We also
demonstrate the capabilities of this methodology by applying it to model the
transport and capture of a cylindrical blood clot in an inferior vena cava
filter.
| no_new_dataset | 0.710666 |
2205.11722 | Rajhans Singh | Rajhans Singh (1), Ankita Shukla (1), Pavan Turaga (1) ((1) Arizona
State University) | Improving Shape Awareness and Interpretability in Deep Networks Using
Geometric Moments | Accepted at CVPR 2023 Workshop: Deep Learning for Geometric Computing | null | null | null | cs.CV | http://creativecommons.org/licenses/by/4.0/ | Deep networks for image classification often rely more on texture information
than object shape. While efforts have been made to make deep-models
shape-aware, it is often difficult to make such models simple, interpretable,
or rooted in known mathematical definitions of shape. This paper presents a
deep-learning model inspired by geometric moments, a classically well
understood approach to measure shape-related properties. The proposed method
consists of a trainable network for generating coordinate bases and affine
parameters for making the features geometrically invariant yet in a
task-specific manner. The proposed model improves the final feature's
interpretation. We demonstrate the effectiveness of our method on standard
image classification datasets. The proposed model achieves higher
classification performance compared to the baseline and standard ResNet models
while substantially improving interpretability.
| [
{
"version": "v1",
"created": "Tue, 24 May 2022 02:08:05 GMT"
},
{
"version": "v2",
"created": "Mon, 22 May 2023 22:02:03 GMT"
}
] | 2023-05-24T00:00:00 | [
[
"Singh",
"Rajhans",
""
],
[
"Shukla",
"Ankita",
""
],
[
"Turaga",
"Pavan",
""
]
] | TITLE: Improving Shape Awareness and Interpretability in Deep Networks Using
Geometric Moments
ABSTRACT: Deep networks for image classification often rely more on texture information
than object shape. While efforts have been made to make deep-models
shape-aware, it is often difficult to make such models simple, interpretable,
or rooted in known mathematical definitions of shape. This paper presents a
deep-learning model inspired by geometric moments, a classically well
understood approach to measure shape-related properties. The proposed method
consists of a trainable network for generating coordinate bases and affine
parameters for making the features geometrically invariant yet in a
task-specific manner. The proposed model improves the final feature's
interpretation. We demonstrate the effectiveness of our method on standard
image classification datasets. The proposed model achieves higher
classification performance compared to the baseline and standard ResNet models
while substantially improving interpretability.
| no_new_dataset | 0.711049 |
2206.00853 | Insung Kong | Insung Kong, Dongyoon Yang, Jongjin Lee, Ilsang Ohn, Yongdai Kim | Masked Bayesian Neural Networks : Computation and Optimality | I will change to another file | null | null | null | stat.ML cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | As data size and computing power increase, the architectures of deep neural
networks (DNNs) have been getting more complex and huge, and thus there is a
growing need to simplify such complex and huge DNNs. In this paper, we propose
a novel sparse Bayesian neural network (BNN) which searches a good DNN with an
appropriate complexity. We employ the masking variables at each node which can
turn off some nodes according to the posterior distribution to yield a nodewise
sparse DNN. We devise a prior distribution such that the posterior distribution
has theoretical optimalities (i.e. minimax optimality and adaptiveness), and
develop an efficient MCMC algorithm. By analyzing several benchmark datasets,
we illustrate that the proposed BNN performs well compared to other existing
methods in the sense that it discovers well condensed DNN architectures with
similar prediction accuracy and uncertainty quantification compared to large
DNNs.
| [
{
"version": "v1",
"created": "Thu, 2 Jun 2022 02:59:55 GMT"
},
{
"version": "v2",
"created": "Tue, 23 May 2023 05:40:32 GMT"
}
] | 2023-05-24T00:00:00 | [
[
"Kong",
"Insung",
""
],
[
"Yang",
"Dongyoon",
""
],
[
"Lee",
"Jongjin",
""
],
[
"Ohn",
"Ilsang",
""
],
[
"Kim",
"Yongdai",
""
]
] | TITLE: Masked Bayesian Neural Networks : Computation and Optimality
ABSTRACT: As data size and computing power increase, the architectures of deep neural
networks (DNNs) have been getting more complex and huge, and thus there is a
growing need to simplify such complex and huge DNNs. In this paper, we propose
a novel sparse Bayesian neural network (BNN) which searches a good DNN with an
appropriate complexity. We employ the masking variables at each node which can
turn off some nodes according to the posterior distribution to yield a nodewise
sparse DNN. We devise a prior distribution such that the posterior distribution
has theoretical optimalities (i.e. minimax optimality and adaptiveness), and
develop an efficient MCMC algorithm. By analyzing several benchmark datasets,
we illustrate that the proposed BNN performs well compared to other existing
methods in the sense that it discovers well condensed DNN architectures with
similar prediction accuracy and uncertainty quantification compared to large
DNNs.
| no_new_dataset | 0.712657 |
2206.01205 | Joel Mathew | Kavitha Raju, Anjaly V, Ryan Lish, Joel Mathew | Snow Mountain: Dataset of Audio Recordings of The Bible in Low Resource
Languages | See dataset at
https://huggingface.co/datasets/bridgeconn/snow-mountain | null | null | null | eess.AS cs.LG cs.SD | http://creativecommons.org/licenses/by-sa/4.0/ | Automatic Speech Recognition (ASR) has increasing utility in the modern
world. There are a many ASR models available for languages with large amounts
of training data like English. However, low-resource languages are poorly
represented. In response we create and release an open-licensed and formatted
dataset of audio recordings of the Bible in low-resource northern Indian
languages. We setup multiple experimental splits and train and analyze two
competitive ASR models to serve as the baseline for future research using this
data.
| [
{
"version": "v1",
"created": "Wed, 1 Jun 2022 18:22:01 GMT"
},
{
"version": "v2",
"created": "Tue, 23 May 2023 05:58:35 GMT"
}
] | 2023-05-24T00:00:00 | [
[
"Raju",
"Kavitha",
""
],
[
"V",
"Anjaly",
""
],
[
"Lish",
"Ryan",
""
],
[
"Mathew",
"Joel",
""
]
] | TITLE: Snow Mountain: Dataset of Audio Recordings of The Bible in Low Resource
Languages
ABSTRACT: Automatic Speech Recognition (ASR) has increasing utility in the modern
world. There are a many ASR models available for languages with large amounts
of training data like English. However, low-resource languages are poorly
represented. In response we create and release an open-licensed and formatted
dataset of audio recordings of the Bible in low-resource northern Indian
languages. We setup multiple experimental splits and train and analyze two
competitive ASR models to serve as the baseline for future research using this
data.
| new_dataset | 0.712245 |
2206.11815 | Nikolay Arefyev | Nikolay Arefyev, Boris Sheludko, Alexander Podolskiy, Alexander
Panchenko | Always Keep your Target in Mind: Studying Semantics and Improving
Performance of Neural Lexical Substitution | arXiv admin note: text overlap with arXiv:2006.00031 | Proceedings of the 28th International Conference on Computational
Linguistics, pages 1242-1255, Barcelona, Spain (Online). International
Committee on Computational Linguistics. 2020 | 10.18653/v1/2020.coling-main.107 | null | cs.CL | http://creativecommons.org/licenses/by/4.0/ | Lexical substitution, i.e. generation of plausible words that can replace a
particular target word in a given context, is an extremely powerful technology
that can be used as a backbone of various NLP applications, including word
sense induction and disambiguation, lexical relation extraction, data
augmentation, etc. In this paper, we present a large-scale comparative study of
lexical substitution methods employing both rather old and most recent language
and masked language models (LMs and MLMs), such as context2vec, ELMo, BERT,
RoBERTa, XLNet. We show that already competitive results achieved by SOTA
LMs/MLMs can be further substantially improved if information about the target
word is injected properly. Several existing and new target word injection
methods are compared for each LM/MLM using both intrinsic evaluation on lexical
substitution datasets and extrinsic evaluation on word sense induction (WSI)
datasets. On two WSI datasets we obtain new SOTA results. Besides, we analyze
the types of semantic relations between target words and their substitutes
generated by different models or given by annotators.
| [
{
"version": "v1",
"created": "Tue, 7 Jun 2022 16:16:19 GMT"
}
] | 2023-05-24T00:00:00 | [
[
"Arefyev",
"Nikolay",
""
],
[
"Sheludko",
"Boris",
""
],
[
"Podolskiy",
"Alexander",
""
],
[
"Panchenko",
"Alexander",
""
]
] | TITLE: Always Keep your Target in Mind: Studying Semantics and Improving
Performance of Neural Lexical Substitution
ABSTRACT: Lexical substitution, i.e. generation of plausible words that can replace a
particular target word in a given context, is an extremely powerful technology
that can be used as a backbone of various NLP applications, including word
sense induction and disambiguation, lexical relation extraction, data
augmentation, etc. In this paper, we present a large-scale comparative study of
lexical substitution methods employing both rather old and most recent language
and masked language models (LMs and MLMs), such as context2vec, ELMo, BERT,
RoBERTa, XLNet. We show that already competitive results achieved by SOTA
LMs/MLMs can be further substantially improved if information about the target
word is injected properly. Several existing and new target word injection
methods are compared for each LM/MLM using both intrinsic evaluation on lexical
substitution datasets and extrinsic evaluation on word sense induction (WSI)
datasets. On two WSI datasets we obtain new SOTA results. Besides, we analyze
the types of semantic relations between target words and their substitutes
generated by different models or given by annotators.
| no_new_dataset | 0.706766 |
2206.14477 | Bin Liu | Lele Wang, Bin Liu | Adversarial Ensemble Training by Jointly Learning Label Dependencies and
Member Models | This paper has been accepted by 19th Inter. Conf. on Intelligent
Computing (ICIC 2023) | null | null | null | cs.LG cs.CR | http://creativecommons.org/licenses/by-nc-nd/4.0/ | Training an ensemble of diverse sub-models has been empirically demonstrated
as an effective strategy for improving the adversarial robustness of deep
neural networks. However, current ensemble training methods for image
recognition typically encode image labels using one-hot vectors, which overlook
dependency relationships between the labels. In this paper, we propose a novel
adversarial en-semble training approach that jointly learns the label
dependencies and member models. Our approach adaptively exploits the learned
label dependencies to pro-mote diversity among the member models. We evaluate
our approach on widely used datasets including MNIST, FashionMNIST, and
CIFAR-10, and show that it achieves superior robustness against black-box
attacks compared to state-of-the-art methods. Our code is available at
https://github.com/ZJLAB-AMMI/LSD.
| [
{
"version": "v1",
"created": "Wed, 29 Jun 2022 09:07:59 GMT"
},
{
"version": "v2",
"created": "Mon, 4 Jul 2022 15:39:44 GMT"
},
{
"version": "v3",
"created": "Tue, 23 May 2023 03:35:34 GMT"
}
] | 2023-05-24T00:00:00 | [
[
"Wang",
"Lele",
""
],
[
"Liu",
"Bin",
""
]
] | TITLE: Adversarial Ensemble Training by Jointly Learning Label Dependencies and
Member Models
ABSTRACT: Training an ensemble of diverse sub-models has been empirically demonstrated
as an effective strategy for improving the adversarial robustness of deep
neural networks. However, current ensemble training methods for image
recognition typically encode image labels using one-hot vectors, which overlook
dependency relationships between the labels. In this paper, we propose a novel
adversarial en-semble training approach that jointly learns the label
dependencies and member models. Our approach adaptively exploits the learned
label dependencies to pro-mote diversity among the member models. We evaluate
our approach on widely used datasets including MNIST, FashionMNIST, and
CIFAR-10, and show that it achieves superior robustness against black-box
attacks compared to state-of-the-art methods. Our code is available at
https://github.com/ZJLAB-AMMI/LSD.
| no_new_dataset | 0.711663 |
2206.14990 | Rongjie Lai | Han Huang and Jiajia Yu and Jie Chen and Rongjie Lai | Bridging Mean-Field Games and Normalizing Flows with Trajectory
Regularization | 36 pages, 22 figures, 9 tables | null | 10.1016/j.jcp.2023.112155 | null | math.OC cs.LG cs.MA | http://creativecommons.org/licenses/by-nc-nd/4.0/ | Mean-field games (MFGs) are a modeling framework for systems with a large
number of interacting agents. They have applications in economics, finance, and
game theory. Normalizing flows (NFs) are a family of deep generative models
that compute data likelihoods by using an invertible mapping, which is
typically parameterized by using neural networks. They are useful for density
modeling and data generation. While active research has been conducted on both
models, few noted the relationship between the two. In this work, we unravel
the connections between MFGs and NFs by contextualizing the training of an NF
as solving the MFG. This is achieved by reformulating the MFG problem in terms
of agent trajectories and parameterizing a discretization of the resulting MFG
with flow architectures. With this connection, we explore two research
directions. First, we employ expressive NF architectures to accurately solve
high-dimensional MFGs, sidestepping the curse of dimensionality in traditional
numerical methods. Compared with other deep learning approaches, our
trajectory-based formulation encodes the continuity equation in the neural
network, resulting in a better approximation of the population dynamics.
Second, we regularize the training of NFs with transport costs and show the
effectiveness on controlling the model's Lipschitz bound, resulting in better
generalization performance. We demonstrate numerical results through
comprehensive experiments on a variety of synthetic and real-life datasets.
| [
{
"version": "v1",
"created": "Thu, 30 Jun 2022 02:44:39 GMT"
}
] | 2023-05-24T00:00:00 | [
[
"Huang",
"Han",
""
],
[
"Yu",
"Jiajia",
""
],
[
"Chen",
"Jie",
""
],
[
"Lai",
"Rongjie",
""
]
] | TITLE: Bridging Mean-Field Games and Normalizing Flows with Trajectory
Regularization
ABSTRACT: Mean-field games (MFGs) are a modeling framework for systems with a large
number of interacting agents. They have applications in economics, finance, and
game theory. Normalizing flows (NFs) are a family of deep generative models
that compute data likelihoods by using an invertible mapping, which is
typically parameterized by using neural networks. They are useful for density
modeling and data generation. While active research has been conducted on both
models, few noted the relationship between the two. In this work, we unravel
the connections between MFGs and NFs by contextualizing the training of an NF
as solving the MFG. This is achieved by reformulating the MFG problem in terms
of agent trajectories and parameterizing a discretization of the resulting MFG
with flow architectures. With this connection, we explore two research
directions. First, we employ expressive NF architectures to accurately solve
high-dimensional MFGs, sidestepping the curse of dimensionality in traditional
numerical methods. Compared with other deep learning approaches, our
trajectory-based formulation encodes the continuity equation in the neural
network, resulting in a better approximation of the population dynamics.
Second, we regularize the training of NFs with transport costs and show the
effectiveness on controlling the model's Lipschitz bound, resulting in better
generalization performance. We demonstrate numerical results through
comprehensive experiments on a variety of synthetic and real-life datasets.
| no_new_dataset | 0.709189 |
2207.07759 | Qi Chang | Qi Chang, Danish Ahmad, Jennifer Toth, Rebecca Bascom, William E.
Higgins | ESFPNet: efficient deep learning architecture for real-time lesion
segmentation in autofluorescence bronchoscopic video | SPIE 2023 drafts update | null | 10.1117/12.2647897 | null | eess.IV cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Lung cancer tends to be detected at an advanced stage, resulting in a high
patient mortality rate. Thus, much recent research has focused on early disease
detection Bronchoscopy is the procedure of choice for an effective noninvasive
way of detecting early manifestations (bronchial lesions) of lung cancer. In
particular, autofluorescence bronchoscopy (AFB) discriminates the
autofluorescence properties of normal (green) and diseased tissue (reddish
brown) with different colors. Because recent studies show AFB's high
sensitivity in searching lesions, it has become a potentially pivotal method in
bronchoscopic airway exams. Unfortunately, manual inspection of AFB video is
extremely tedious and error prone, while limited effort has been expended
toward potentially more robust automatic AFB lesion analysis. We propose a
real-time (processing throughput of 27 frames/sec) deep-learning architecture
dubbed ESFPNet for accurate segmentation and robust detection of bronchial
lesions in AFB video streams. The architecture features an encoder structure
that exploits pretrained Mix Transformer (MiT) encoders and an efficient
stage-wise feature pyramid (ESFP) decoder structure. Segmentation results from
the AFB airway-exam videos of 20 lung cancer patients indicate that our
approach gives a mean Dice index = 0.756 and an average Intersection of Union =
0.624, results that are superior to those generated by other recent
architectures. Thus, ESFPNet gives the physician a potential tool for confident
real-time lesion segmentation and detection during a live bronchoscopic airway
exam. Moreover, our model shows promising potential applicability to other
domains, as evidenced by its state-of-the-art (SOTA) performance on the
CVC-ClinicDB, ETIS-LaribPolypDB datasets, and superior performance on the
Kvasir, CVC-ColonDB datasets.
| [
{
"version": "v1",
"created": "Fri, 15 Jul 2022 21:21:26 GMT"
},
{
"version": "v2",
"created": "Thu, 25 Aug 2022 20:28:21 GMT"
},
{
"version": "v3",
"created": "Fri, 9 Dec 2022 04:21:41 GMT"
}
] | 2023-05-24T00:00:00 | [
[
"Chang",
"Qi",
""
],
[
"Ahmad",
"Danish",
""
],
[
"Toth",
"Jennifer",
""
],
[
"Bascom",
"Rebecca",
""
],
[
"Higgins",
"William E.",
""
]
] | TITLE: ESFPNet: efficient deep learning architecture for real-time lesion
segmentation in autofluorescence bronchoscopic video
ABSTRACT: Lung cancer tends to be detected at an advanced stage, resulting in a high
patient mortality rate. Thus, much recent research has focused on early disease
detection Bronchoscopy is the procedure of choice for an effective noninvasive
way of detecting early manifestations (bronchial lesions) of lung cancer. In
particular, autofluorescence bronchoscopy (AFB) discriminates the
autofluorescence properties of normal (green) and diseased tissue (reddish
brown) with different colors. Because recent studies show AFB's high
sensitivity in searching lesions, it has become a potentially pivotal method in
bronchoscopic airway exams. Unfortunately, manual inspection of AFB video is
extremely tedious and error prone, while limited effort has been expended
toward potentially more robust automatic AFB lesion analysis. We propose a
real-time (processing throughput of 27 frames/sec) deep-learning architecture
dubbed ESFPNet for accurate segmentation and robust detection of bronchial
lesions in AFB video streams. The architecture features an encoder structure
that exploits pretrained Mix Transformer (MiT) encoders and an efficient
stage-wise feature pyramid (ESFP) decoder structure. Segmentation results from
the AFB airway-exam videos of 20 lung cancer patients indicate that our
approach gives a mean Dice index = 0.756 and an average Intersection of Union =
0.624, results that are superior to those generated by other recent
architectures. Thus, ESFPNet gives the physician a potential tool for confident
real-time lesion segmentation and detection during a live bronchoscopic airway
exam. Moreover, our model shows promising potential applicability to other
domains, as evidenced by its state-of-the-art (SOTA) performance on the
CVC-ClinicDB, ETIS-LaribPolypDB datasets, and superior performance on the
Kvasir, CVC-ColonDB datasets.
| no_new_dataset | 0.709818 |
2208.04726 | Lahav Lipson | Zachary Teed, Lahav Lipson and Jia Deng | Deep Patch Visual Odometry | null | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We propose Deep Patch Visual Odometry (DPVO), a new deep learning system for
monocular Visual Odometry (VO). DPVO uses a novel recurrent network
architecture designed for tracking image patches across time. Recent approaches
to VO have significantly improved the state-of-the-art accuracy by using deep
networks to predict dense flow between video frames. However, using dense flow
incurs a large computational cost, making these previous methods impractical
for many use cases. Despite this, it has been assumed that dense flow is
important as it provides additional redundancy against incorrect matches. DPVO
disproves this assumption, showing that it is possible to get the best accuracy
and efficiency by exploiting the advantages of sparse patch-based matching over
dense flow. DPVO introduces a novel recurrent update operator for patch based
correspondence coupled with differentiable bundle adjustment. On Standard
benchmarks, DPVO outperforms all prior work, including the learning-based
state-of-the-art VO-system (DROID) using a third of the memory while running 3x
faster on average. Code is available at https://github.com/princeton-vl/DPVO
| [
{
"version": "v1",
"created": "Mon, 8 Aug 2022 04:47:38 GMT"
},
{
"version": "v2",
"created": "Tue, 23 May 2023 17:59:29 GMT"
}
] | 2023-05-24T00:00:00 | [
[
"Teed",
"Zachary",
""
],
[
"Lipson",
"Lahav",
""
],
[
"Deng",
"Jia",
""
]
] | TITLE: Deep Patch Visual Odometry
ABSTRACT: We propose Deep Patch Visual Odometry (DPVO), a new deep learning system for
monocular Visual Odometry (VO). DPVO uses a novel recurrent network
architecture designed for tracking image patches across time. Recent approaches
to VO have significantly improved the state-of-the-art accuracy by using deep
networks to predict dense flow between video frames. However, using dense flow
incurs a large computational cost, making these previous methods impractical
for many use cases. Despite this, it has been assumed that dense flow is
important as it provides additional redundancy against incorrect matches. DPVO
disproves this assumption, showing that it is possible to get the best accuracy
and efficiency by exploiting the advantages of sparse patch-based matching over
dense flow. DPVO introduces a novel recurrent update operator for patch based
correspondence coupled with differentiable bundle adjustment. On Standard
benchmarks, DPVO outperforms all prior work, including the learning-based
state-of-the-art VO-system (DROID) using a third of the memory while running 3x
faster on average. Code is available at https://github.com/princeton-vl/DPVO
| no_new_dataset | 0.711425 |
2209.02132 | Chengyin Hu | Chengyin Hu, Weiwen Shi | Impact of Scaled Image on Robustness of Deep Neural Networks | null | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Deep neural networks (DNNs) have been widely used in computer vision tasks
like image classification, object detection and segmentation. Whereas recent
studies have shown their vulnerability to manual digital perturbations or
distortion in the input images. The accuracy of the networks is remarkably
influenced by the data distribution of their training dataset. Scaling the raw
images creates out-of-distribution data, which makes it a possible adversarial
attack to fool the networks. In this work, we propose a Scaling-distortion
dataset ImageNet-CS by Scaling a subset of the ImageNet Challenge dataset by
different multiples. The aim of our work is to study the impact of scaled
images on the performance of advanced DNNs. We perform experiments on several
state-of-the-art deep neural network architectures on the proposed ImageNet-CS,
and the results show a significant positive correlation between scaling size
and accuracy decline. Moreover, based on ResNet50 architecture, we demonstrate
some tests on the performance of recent proposed robust training techniques and
strategies like Augmix, Revisiting and Normalizer Free on our proposed
ImageNet-CS. Experiment results have shown that these robust training
techniques can improve networks' robustness to scaling transformation.
| [
{
"version": "v1",
"created": "Fri, 2 Sep 2022 08:06:58 GMT"
},
{
"version": "v2",
"created": "Tue, 23 May 2023 15:46:32 GMT"
}
] | 2023-05-24T00:00:00 | [
[
"Hu",
"Chengyin",
""
],
[
"Shi",
"Weiwen",
""
]
] | TITLE: Impact of Scaled Image on Robustness of Deep Neural Networks
ABSTRACT: Deep neural networks (DNNs) have been widely used in computer vision tasks
like image classification, object detection and segmentation. Whereas recent
studies have shown their vulnerability to manual digital perturbations or
distortion in the input images. The accuracy of the networks is remarkably
influenced by the data distribution of their training dataset. Scaling the raw
images creates out-of-distribution data, which makes it a possible adversarial
attack to fool the networks. In this work, we propose a Scaling-distortion
dataset ImageNet-CS by Scaling a subset of the ImageNet Challenge dataset by
different multiples. The aim of our work is to study the impact of scaled
images on the performance of advanced DNNs. We perform experiments on several
state-of-the-art deep neural network architectures on the proposed ImageNet-CS,
and the results show a significant positive correlation between scaling size
and accuracy decline. Moreover, based on ResNet50 architecture, we demonstrate
some tests on the performance of recent proposed robust training techniques and
strategies like Augmix, Revisiting and Normalizer Free on our proposed
ImageNet-CS. Experiment results have shown that these robust training
techniques can improve networks' robustness to scaling transformation.
| no_new_dataset | 0.708988 |
2209.02832 | Chengyin Hu | Chengyin Hu, Weiwen Shi | Impact of Colour Variation on Robustness of Deep Neural Networks | arXiv admin note: substantial text overlap with arXiv:2209.02132 | null | null | null | cs.CV cs.LG eess.IV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Deep neural networks (DNNs) have have shown state-of-the-art performance for
computer vision applications like image classification, segmentation and object
detection. Whereas recent advances have shown their vulnerability to manual
digital perturbations in the input data, namely adversarial attacks. The
accuracy of the networks is significantly affected by the data distribution of
their training dataset. Distortions or perturbations on color space of input
images generates out-of-distribution data, which make networks more likely to
misclassify them. In this work, we propose a color-variation dataset by
distorting their RGB color on a subset of the ImageNet with 27 different
combinations. The aim of our work is to study the impact of color variation on
the performance of DNNs. We perform experiments on several state-of-the-art DNN
architectures on the proposed dataset, and the result shows a significant
correlation between color variation and loss of accuracy. Furthermore, based on
the ResNet50 architecture, we demonstrate some experiments of the performance
of recently proposed robust training techniques and strategies, such as Augmix,
revisit, and free normalizer, on our proposed dataset. Experimental results
indicate that these robust training techniques can improve the robustness of
deep networks to color variation.
| [
{
"version": "v1",
"created": "Fri, 2 Sep 2022 08:16:04 GMT"
},
{
"version": "v2",
"created": "Tue, 23 May 2023 15:43:28 GMT"
}
] | 2023-05-24T00:00:00 | [
[
"Hu",
"Chengyin",
""
],
[
"Shi",
"Weiwen",
""
]
] | TITLE: Impact of Colour Variation on Robustness of Deep Neural Networks
ABSTRACT: Deep neural networks (DNNs) have have shown state-of-the-art performance for
computer vision applications like image classification, segmentation and object
detection. Whereas recent advances have shown their vulnerability to manual
digital perturbations in the input data, namely adversarial attacks. The
accuracy of the networks is significantly affected by the data distribution of
their training dataset. Distortions or perturbations on color space of input
images generates out-of-distribution data, which make networks more likely to
misclassify them. In this work, we propose a color-variation dataset by
distorting their RGB color on a subset of the ImageNet with 27 different
combinations. The aim of our work is to study the impact of color variation on
the performance of DNNs. We perform experiments on several state-of-the-art DNN
architectures on the proposed dataset, and the result shows a significant
correlation between color variation and loss of accuracy. Furthermore, based on
the ResNet50 architecture, we demonstrate some experiments of the performance
of recently proposed robust training techniques and strategies, such as Augmix,
revisit, and free normalizer, on our proposed dataset. Experimental results
indicate that these robust training techniques can improve the robustness of
deep networks to color variation.
| new_dataset | 0.715977 |
2209.07634 | Qingyang Wu | Qingyang Wu and Zhou Yu | Stateful Memory-Augmented Transformers for Efficient Dialogue Modeling | null | null | null | null | cs.CL | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Transformer encoder-decoder models have achieved great performance in
dialogue generation tasks, however, their inability to process long dialogue
history often leads to truncation of the context To address this problem, we
propose a novel memory-augmented transformer that is compatible with existing
pre-trained encoder-decoder models and enables efficient preservation of the
dialogue history information. By incorporating a separate memory module
alongside the pre-trained transformer, the model can effectively interchange
information between the memory states and the current input context. We
evaluate our model on three dialogue datasets and two language modeling
datasets. Experimental results show that our method has achieved superior
efficiency and performance compared to other pre-trained Transformer baselines.
| [
{
"version": "v1",
"created": "Thu, 15 Sep 2022 22:37:22 GMT"
},
{
"version": "v2",
"created": "Tue, 23 May 2023 05:59:06 GMT"
}
] | 2023-05-24T00:00:00 | [
[
"Wu",
"Qingyang",
""
],
[
"Yu",
"Zhou",
""
]
] | TITLE: Stateful Memory-Augmented Transformers for Efficient Dialogue Modeling
ABSTRACT: Transformer encoder-decoder models have achieved great performance in
dialogue generation tasks, however, their inability to process long dialogue
history often leads to truncation of the context To address this problem, we
propose a novel memory-augmented transformer that is compatible with existing
pre-trained encoder-decoder models and enables efficient preservation of the
dialogue history information. By incorporating a separate memory module
alongside the pre-trained transformer, the model can effectively interchange
information between the memory states and the current input context. We
evaluate our model on three dialogue datasets and two language modeling
datasets. Experimental results show that our method has achieved superior
efficiency and performance compared to other pre-trained Transformer baselines.
| no_new_dataset | 0.711055 |
2209.11820 | Boris Ivanovic | Boris Ivanovic, James Harrison, Marco Pavone | Expanding the Deployment Envelope of Behavior Prediction via Adaptive
Meta-Learning | 12 pages, 13 figures, 2 tables. ICRA 2023 | null | null | null | cs.LG cs.CV cs.RO | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Learning-based behavior prediction methods are increasingly being deployed in
real-world autonomous systems, e.g., in fleets of self-driving vehicles, which
are beginning to commercially operate in major cities across the world. Despite
their advancements, however, the vast majority of prediction systems are
specialized to a set of well-explored geographic regions or operational design
domains, complicating deployment to additional cities, countries, or
continents. Towards this end, we present a novel method for efficiently
adapting behavior prediction models to new environments. Our approach leverages
recent advances in meta-learning, specifically Bayesian regression, to augment
existing behavior prediction models with an adaptive layer that enables
efficient domain transfer via offline fine-tuning, online adaptation, or both.
Experiments across multiple real-world datasets demonstrate that our method can
efficiently adapt to a variety of unseen environments.
| [
{
"version": "v1",
"created": "Fri, 23 Sep 2022 19:23:07 GMT"
},
{
"version": "v2",
"created": "Tue, 25 Oct 2022 17:55:34 GMT"
},
{
"version": "v3",
"created": "Wed, 1 Feb 2023 03:25:21 GMT"
},
{
"version": "v4",
"created": "Tue, 23 May 2023 05:37:16 GMT"
}
] | 2023-05-24T00:00:00 | [
[
"Ivanovic",
"Boris",
""
],
[
"Harrison",
"James",
""
],
[
"Pavone",
"Marco",
""
]
] | TITLE: Expanding the Deployment Envelope of Behavior Prediction via Adaptive
Meta-Learning
ABSTRACT: Learning-based behavior prediction methods are increasingly being deployed in
real-world autonomous systems, e.g., in fleets of self-driving vehicles, which
are beginning to commercially operate in major cities across the world. Despite
their advancements, however, the vast majority of prediction systems are
specialized to a set of well-explored geographic regions or operational design
domains, complicating deployment to additional cities, countries, or
continents. Towards this end, we present a novel method for efficiently
adapting behavior prediction models to new environments. Our approach leverages
recent advances in meta-learning, specifically Bayesian regression, to augment
existing behavior prediction models with an adaptive layer that enables
efficient domain transfer via offline fine-tuning, online adaptation, or both.
Experiments across multiple real-world datasets demonstrate that our method can
efficiently adapt to a variety of unseen environments.
| no_new_dataset | 0.70883 |
2209.14734 | Cl\'ement Vignac | Clement Vignac, Igor Krawczuk, Antoine Siraudin, Bohan Wang, Volkan
Cevher, Pascal Frossard | DiGress: Discrete Denoising diffusion for graph generation | 22 pages. Published as a conference paper at ICLR 2023 | International Conference on Learning Representations (ICLR 2023) | null | null | cs.LG | http://creativecommons.org/licenses/by-sa/4.0/ | This work introduces DiGress, a discrete denoising diffusion model for
generating graphs with categorical node and edge attributes. Our model utilizes
a discrete diffusion process that progressively edits graphs with noise,
through the process of adding or removing edges and changing the categories. A
graph transformer network is trained to revert this process, simplifying the
problem of distribution learning over graphs into a sequence of node and edge
classification tasks. We further improve sample quality by introducing a
Markovian noise model that preserves the marginal distribution of node and edge
types during diffusion, and by incorporating auxiliary graph-theoretic
features. A procedure for conditioning the generation on graph-level features
is also proposed. DiGress achieves state-of-the-art performance on molecular
and non-molecular datasets, with up to 3x validity improvement on a planar
graph dataset. It is also the first model to scale to the large GuacaMol
dataset containing 1.3M drug-like molecules without the use of
molecule-specific representations.
| [
{
"version": "v1",
"created": "Thu, 29 Sep 2022 12:55:03 GMT"
},
{
"version": "v2",
"created": "Fri, 28 Oct 2022 17:00:04 GMT"
},
{
"version": "v3",
"created": "Mon, 6 Feb 2023 14:38:00 GMT"
},
{
"version": "v4",
"created": "Tue, 23 May 2023 10:32:08 GMT"
}
] | 2023-05-24T00:00:00 | [
[
"Vignac",
"Clement",
""
],
[
"Krawczuk",
"Igor",
""
],
[
"Siraudin",
"Antoine",
""
],
[
"Wang",
"Bohan",
""
],
[
"Cevher",
"Volkan",
""
],
[
"Frossard",
"Pascal",
""
]
] | TITLE: DiGress: Discrete Denoising diffusion for graph generation
ABSTRACT: This work introduces DiGress, a discrete denoising diffusion model for
generating graphs with categorical node and edge attributes. Our model utilizes
a discrete diffusion process that progressively edits graphs with noise,
through the process of adding or removing edges and changing the categories. A
graph transformer network is trained to revert this process, simplifying the
problem of distribution learning over graphs into a sequence of node and edge
classification tasks. We further improve sample quality by introducing a
Markovian noise model that preserves the marginal distribution of node and edge
types during diffusion, and by incorporating auxiliary graph-theoretic
features. A procedure for conditioning the generation on graph-level features
is also proposed. DiGress achieves state-of-the-art performance on molecular
and non-molecular datasets, with up to 3x validity improvement on a planar
graph dataset. It is also the first model to scale to the large GuacaMol
dataset containing 1.3M drug-like molecules without the use of
molecule-specific representations.
| no_new_dataset | 0.71086 |
2210.01371 | Man Luo | Man Luo, Shashank Jain, Anchit Gupta, Arash Einolghozati, Barlas Oguz,
Debojeet Chatterjee, Xilun Chen, Chitta Baral, Peyman Heidari | A Study on the Efficiency and Generalization of Light Hybrid Retrievers | accepted to ACL23 | null | null | null | cs.IR cs.CL | http://creativecommons.org/licenses/by/4.0/ | Hybrid retrievers can take advantage of both sparse and dense retrievers.
Previous hybrid retrievers leverage indexing-heavy dense retrievers. In this
work, we study "Is it possible to reduce the indexing memory of hybrid
retrievers without sacrificing performance"? Driven by this question, we
leverage an indexing-efficient dense retriever (i.e. DrBoost) and introduce a
LITE retriever that further reduces the memory of DrBoost. LITE is jointly
trained on contrastive learning and knowledge distillation from DrBoost. Then,
we integrate BM25, a sparse retriever, with either LITE or DrBoost to form
light hybrid retrievers. Our Hybrid-LITE retriever saves 13X memory while
maintaining 98.0% performance of the hybrid retriever of BM25 and DPR. In
addition, we study the generalization capacity of our light hybrid retrievers
on out-of-domain dataset and a set of adversarial attacks datasets. Experiments
showcase that light hybrid retrievers achieve better generalization performance
than individual sparse and dense retrievers. Nevertheless, our analysis shows
that there is a large room to improve the robustness of retrievers, suggesting
a new research direction.
| [
{
"version": "v1",
"created": "Tue, 4 Oct 2022 04:22:46 GMT"
},
{
"version": "v2",
"created": "Tue, 23 May 2023 09:45:30 GMT"
}
] | 2023-05-24T00:00:00 | [
[
"Luo",
"Man",
""
],
[
"Jain",
"Shashank",
""
],
[
"Gupta",
"Anchit",
""
],
[
"Einolghozati",
"Arash",
""
],
[
"Oguz",
"Barlas",
""
],
[
"Chatterjee",
"Debojeet",
""
],
[
"Chen",
"Xilun",
""
],
[
"Baral",
"Chitta",
""
],
[
"Heidari",
"Peyman",
""
]
] | TITLE: A Study on the Efficiency and Generalization of Light Hybrid Retrievers
ABSTRACT: Hybrid retrievers can take advantage of both sparse and dense retrievers.
Previous hybrid retrievers leverage indexing-heavy dense retrievers. In this
work, we study "Is it possible to reduce the indexing memory of hybrid
retrievers without sacrificing performance"? Driven by this question, we
leverage an indexing-efficient dense retriever (i.e. DrBoost) and introduce a
LITE retriever that further reduces the memory of DrBoost. LITE is jointly
trained on contrastive learning and knowledge distillation from DrBoost. Then,
we integrate BM25, a sparse retriever, with either LITE or DrBoost to form
light hybrid retrievers. Our Hybrid-LITE retriever saves 13X memory while
maintaining 98.0% performance of the hybrid retriever of BM25 and DPR. In
addition, we study the generalization capacity of our light hybrid retrievers
on out-of-domain dataset and a set of adversarial attacks datasets. Experiments
showcase that light hybrid retrievers achieve better generalization performance
than individual sparse and dense retrievers. Nevertheless, our analysis shows
that there is a large room to improve the robustness of retrievers, suggesting
a new research direction.
| no_new_dataset | 0.704313 |
2210.04325 | Jiannan Xiang | Jiannan Xiang, Zhengzhong Liu, Yucheng Zhou, Eric P. Xing, Zhiting Hu | ASDOT: Any-Shot Data-to-Text Generation with Pretrained Language Models | Findings of EMNLP 2022 | null | null | null | cs.CL cs.AI cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Data-to-text generation is challenging due to the great variety of the input
data in terms of domains (e.g., finance vs sports) or schemata (e.g., diverse
predicates). Recent end-to-end neural methods thus require substantial training
examples to learn to disambiguate and describe the data. Yet, real-world
data-to-text problems often suffer from various data-scarce issues: one may
have access to only a handful of or no training examples, and/or have to rely
on examples in a different domain or schema. To fill this gap, we propose
Any-Shot Data-to-Text (ASDOT), a new approach flexibly applicable to diverse
settings by making efficient use of any given (or no) examples. ASDOT consists
of two steps, data disambiguation and sentence fusion, both of which are
amenable to be solved with off-the-shelf pretrained language models (LMs) with
optional finetuning. In the data disambiguation stage, we employ the prompted
GPT-3 model to understand possibly ambiguous triples from the input data and
convert each into a short sentence with reduced ambiguity. The sentence fusion
stage then uses an LM like T5 to fuse all the resulting sentences into a
coherent paragraph as the final description. We evaluate extensively on various
datasets in different scenarios, including the zero-/few-/full-shot settings,
and generalization to unseen predicates and out-of-domain data. Experimental
results show that ASDOT consistently achieves significant improvement over
baselines, e.g., a 30.81 BLEU gain on the DART dataset under the zero-shot
setting.
| [
{
"version": "v1",
"created": "Sun, 9 Oct 2022 19:17:43 GMT"
},
{
"version": "v2",
"created": "Tue, 11 Oct 2022 03:33:06 GMT"
},
{
"version": "v3",
"created": "Sat, 22 Oct 2022 08:50:44 GMT"
}
] | 2023-05-24T00:00:00 | [
[
"Xiang",
"Jiannan",
""
],
[
"Liu",
"Zhengzhong",
""
],
[
"Zhou",
"Yucheng",
""
],
[
"Xing",
"Eric P.",
""
],
[
"Hu",
"Zhiting",
""
]
] | TITLE: ASDOT: Any-Shot Data-to-Text Generation with Pretrained Language Models
ABSTRACT: Data-to-text generation is challenging due to the great variety of the input
data in terms of domains (e.g., finance vs sports) or schemata (e.g., diverse
predicates). Recent end-to-end neural methods thus require substantial training
examples to learn to disambiguate and describe the data. Yet, real-world
data-to-text problems often suffer from various data-scarce issues: one may
have access to only a handful of or no training examples, and/or have to rely
on examples in a different domain or schema. To fill this gap, we propose
Any-Shot Data-to-Text (ASDOT), a new approach flexibly applicable to diverse
settings by making efficient use of any given (or no) examples. ASDOT consists
of two steps, data disambiguation and sentence fusion, both of which are
amenable to be solved with off-the-shelf pretrained language models (LMs) with
optional finetuning. In the data disambiguation stage, we employ the prompted
GPT-3 model to understand possibly ambiguous triples from the input data and
convert each into a short sentence with reduced ambiguity. The sentence fusion
stage then uses an LM like T5 to fuse all the resulting sentences into a
coherent paragraph as the final description. We evaluate extensively on various
datasets in different scenarios, including the zero-/few-/full-shot settings,
and generalization to unseen predicates and out-of-domain data. Experimental
results show that ASDOT consistently achieves significant improvement over
baselines, e.g., a 30.81 BLEU gain on the DART dataset under the zero-shot
setting.
| no_new_dataset | 0.712664 |
2210.05156 | Hao Fang | Hao Cheng, Hao Fang, Xiaodong Liu, Jianfeng Gao | Task-Aware Specialization for Efficient and Robust Dense Retrieval for
Open-Domain Question Answering | null | null | null | null | cs.CL | http://creativecommons.org/licenses/by/4.0/ | Given its effectiveness on knowledge-intensive natural language processing
tasks, dense retrieval models have become increasingly popular. Specifically,
the de-facto architecture for open-domain question answering uses two
isomorphic encoders that are initialized from the same pretrained model but
separately parameterized for questions and passages. This bi-encoder
architecture is parameter-inefficient in that there is no parameter sharing
between encoders. Further, recent studies show that such dense retrievers
underperform BM25 in various settings. We thus propose a new architecture,
Task-aware Specialization for dense Retrieval (TASER), which enables parameter
sharing by interleaving shared and specialized blocks in a single encoder. Our
experiments on five question answering datasets show that TASER can achieve
superior accuracy, surpassing BM25, while using about 60% of the parameters as
bi-encoder dense retrievers. In out-of-domain evaluations, TASER is also
empirically more robust than bi-encoder dense retrievers. Our code is available
at https://github.com/microsoft/taser.
| [
{
"version": "v1",
"created": "Tue, 11 Oct 2022 05:33:25 GMT"
},
{
"version": "v2",
"created": "Mon, 22 May 2023 20:38:56 GMT"
}
] | 2023-05-24T00:00:00 | [
[
"Cheng",
"Hao",
""
],
[
"Fang",
"Hao",
""
],
[
"Liu",
"Xiaodong",
""
],
[
"Gao",
"Jianfeng",
""
]
] | TITLE: Task-Aware Specialization for Efficient and Robust Dense Retrieval for
Open-Domain Question Answering
ABSTRACT: Given its effectiveness on knowledge-intensive natural language processing
tasks, dense retrieval models have become increasingly popular. Specifically,
the de-facto architecture for open-domain question answering uses two
isomorphic encoders that are initialized from the same pretrained model but
separately parameterized for questions and passages. This bi-encoder
architecture is parameter-inefficient in that there is no parameter sharing
between encoders. Further, recent studies show that such dense retrievers
underperform BM25 in various settings. We thus propose a new architecture,
Task-aware Specialization for dense Retrieval (TASER), which enables parameter
sharing by interleaving shared and specialized blocks in a single encoder. Our
experiments on five question answering datasets show that TASER can achieve
superior accuracy, surpassing BM25, while using about 60% of the parameters as
bi-encoder dense retrievers. In out-of-domain evaluations, TASER is also
empirically more robust than bi-encoder dense retrievers. Our code is available
at https://github.com/microsoft/taser.
| no_new_dataset | 0.709982 |
2210.09015 | Avinash Kori | Avinash Kori, Ben Glocker, Francesca Toni | Explaining Image Classification with Visual Debates | null | null | null | null | cs.CV cs.LG | http://creativecommons.org/licenses/by-nc-nd/4.0/ | An effective way to obtain different perspectives on any given topic is by
conducting a debate, where participants argue for and against the topic. Here,
we propose a novel debate framework for understanding and explaining a
continuous image classifier's reasoning for making a particular prediction by
modeling it as a multiplayer sequential zero-sum debate game. The contrastive
nature of our framework encourages players to learn to put forward diverse
arguments during the debates, picking up the reasoning trails missed by their
opponents and highlighting any uncertainties in the classifier. Specifically,
in our proposed setup, players propose arguments, drawn from the classifier's
discretized latent knowledge, to support or oppose the classifier's decision.
The resulting Visual Debates collect supporting and opposing features from the
discretized latent space of the classifier, serving as explanations for the
internal reasoning of the classifier towards its predictions. We demonstrate
and evaluate (a practical realization of) our Visual Debates on the geometric
SHAPE and MNIST datasets and on the high-resolution animal faces (AFHQ)
dataset, along standard evaluation metrics for explanations (i.e. faithfulness
and completeness) and novel, bespoke metrics for visual debates as explanations
(consensus and split ratio).
| [
{
"version": "v1",
"created": "Mon, 17 Oct 2022 12:35:52 GMT"
},
{
"version": "v2",
"created": "Tue, 23 May 2023 09:58:18 GMT"
}
] | 2023-05-24T00:00:00 | [
[
"Kori",
"Avinash",
""
],
[
"Glocker",
"Ben",
""
],
[
"Toni",
"Francesca",
""
]
] | TITLE: Explaining Image Classification with Visual Debates
ABSTRACT: An effective way to obtain different perspectives on any given topic is by
conducting a debate, where participants argue for and against the topic. Here,
we propose a novel debate framework for understanding and explaining a
continuous image classifier's reasoning for making a particular prediction by
modeling it as a multiplayer sequential zero-sum debate game. The contrastive
nature of our framework encourages players to learn to put forward diverse
arguments during the debates, picking up the reasoning trails missed by their
opponents and highlighting any uncertainties in the classifier. Specifically,
in our proposed setup, players propose arguments, drawn from the classifier's
discretized latent knowledge, to support or oppose the classifier's decision.
The resulting Visual Debates collect supporting and opposing features from the
discretized latent space of the classifier, serving as explanations for the
internal reasoning of the classifier towards its predictions. We demonstrate
and evaluate (a practical realization of) our Visual Debates on the geometric
SHAPE and MNIST datasets and on the high-resolution animal faces (AFHQ)
dataset, along standard evaluation metrics for explanations (i.e. faithfulness
and completeness) and novel, bespoke metrics for visual debates as explanations
(consensus and split ratio).
| no_new_dataset | 0.711249 |
2210.12386 | Joaquim Ortiz-Haro | Joaquim Ortiz-Haro, Jung-Su Ha, Danny Driess, Erez Karpas, Marc
Toussaint | Learning Feasibility of Factored Nonlinear Programs in Robotic
Manipulation Planning | Submitted to ICRA 2023 | null | null | null | cs.RO | http://creativecommons.org/licenses/by/4.0/ | A factored Nonlinear Program (Factored-NLP) explicitly models the
dependencies between a set of continuous variables and nonlinear constraints,
providing an expressive formulation for relevant robotics problems such as
manipulation planning or simultaneous localization and mapping. When the
problem is over-constrained or infeasible, a fundamental issue is to detect a
minimal subset of variables and constraints that are infeasible. Previous
approaches require solving several nonlinear programs, incrementally adding and
removing constraints, and are thus computationally expensive. In this paper, we
propose a graph neural architecture that predicts which variables and
constraints are jointly infeasible. The model is trained with a dataset of
labeled subgraphs of Factored-NLPs, and importantly, can make useful
predictions on larger factored nonlinear programs than the ones seen during
training. We evaluate our approach in robotic manipulation planning, where our
model is able to generalize to longer manipulation sequences involving more
objects and robots, and different geometric environments. The experiments show
that the learned model accelerates general algorithms for conflict extraction
(by a factor of 50) and heuristic algorithms that exploit expert knowledge (by
a factor of 4).
| [
{
"version": "v1",
"created": "Sat, 22 Oct 2022 08:38:00 GMT"
},
{
"version": "v2",
"created": "Tue, 23 May 2023 10:44:42 GMT"
}
] | 2023-05-24T00:00:00 | [
[
"Ortiz-Haro",
"Joaquim",
""
],
[
"Ha",
"Jung-Su",
""
],
[
"Driess",
"Danny",
""
],
[
"Karpas",
"Erez",
""
],
[
"Toussaint",
"Marc",
""
]
] | TITLE: Learning Feasibility of Factored Nonlinear Programs in Robotic
Manipulation Planning
ABSTRACT: A factored Nonlinear Program (Factored-NLP) explicitly models the
dependencies between a set of continuous variables and nonlinear constraints,
providing an expressive formulation for relevant robotics problems such as
manipulation planning or simultaneous localization and mapping. When the
problem is over-constrained or infeasible, a fundamental issue is to detect a
minimal subset of variables and constraints that are infeasible. Previous
approaches require solving several nonlinear programs, incrementally adding and
removing constraints, and are thus computationally expensive. In this paper, we
propose a graph neural architecture that predicts which variables and
constraints are jointly infeasible. The model is trained with a dataset of
labeled subgraphs of Factored-NLPs, and importantly, can make useful
predictions on larger factored nonlinear programs than the ones seen during
training. We evaluate our approach in robotic manipulation planning, where our
model is able to generalize to longer manipulation sequences involving more
objects and robots, and different geometric environments. The experiments show
that the learned model accelerates general algorithms for conflict extraction
(by a factor of 50) and heuristic algorithms that exploit expert knowledge (by
a factor of 4).
| no_new_dataset | 0.709265 |
2210.14564 | Myunghun Jung | Myunghun Jung, Hoirin Kim | AdaMS: Deep Metric Learning with Adaptive Margin and Adaptive Scale for
Acoustic Word Discrimination | INTERSPEECH 2023 | null | null | null | eess.AS cs.IR cs.LG cs.SD | http://creativecommons.org/licenses/by-nc-nd/4.0/ | Many recent loss functions in deep metric learning are expressed with
logarithmic and exponential forms, and they involve margin and scale as
essential hyper-parameters. Since each data class has an intrinsic
characteristic, several previous works have tried to learn embedding space
close to the real distribution by introducing adaptive margins. However, there
was no work on adaptive scales at all. We argue that both margin and scale
should be adaptively adjustable during the training. In this paper, we propose
a method called Adaptive Margin and Scale (AdaMS), where hyper-parameters of
margin and scale are replaced with learnable parameters of adaptive margins and
adaptive scales for each class. Our method is evaluated on Wall Street Journal
dataset, and we achieve outperforming results for word discrimination tasks.
| [
{
"version": "v1",
"created": "Wed, 26 Oct 2022 08:53:31 GMT"
},
{
"version": "v2",
"created": "Tue, 23 May 2023 06:42:56 GMT"
}
] | 2023-05-24T00:00:00 | [
[
"Jung",
"Myunghun",
""
],
[
"Kim",
"Hoirin",
""
]
] | TITLE: AdaMS: Deep Metric Learning with Adaptive Margin and Adaptive Scale for
Acoustic Word Discrimination
ABSTRACT: Many recent loss functions in deep metric learning are expressed with
logarithmic and exponential forms, and they involve margin and scale as
essential hyper-parameters. Since each data class has an intrinsic
characteristic, several previous works have tried to learn embedding space
close to the real distribution by introducing adaptive margins. However, there
was no work on adaptive scales at all. We argue that both margin and scale
should be adaptively adjustable during the training. In this paper, we propose
a method called Adaptive Margin and Scale (AdaMS), where hyper-parameters of
margin and scale are replaced with learnable parameters of adaptive margins and
adaptive scales for each class. Our method is evaluated on Wall Street Journal
dataset, and we achieve outperforming results for word discrimination tasks.
| no_new_dataset | 0.711036 |
2211.01335 | Junyang Lin | An Yang, Junshu Pan, Junyang Lin, Rui Men, Yichang Zhang, Jingren
Zhou, Chang Zhou | Chinese CLIP: Contrastive Vision-Language Pretraining in Chinese | null | null | null | null | cs.CV cs.CL | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | The tremendous success of CLIP (Radford et al., 2021) has promoted the
research and application of contrastive learning for vision-language
pretraining. In this work, we construct a large-scale dataset of image-text
pairs in Chinese, where most data are retrieved from publicly available
datasets, and we pretrain Chinese CLIP models on the new dataset. We develop 5
Chinese CLIP models of multiple sizes, spanning from 77 to 958 million
parameters. Furthermore, we propose a two-stage pretraining method, where the
model is first trained with the image encoder frozen and then trained with all
parameters being optimized, to achieve enhanced model performance. Our
comprehensive experiments demonstrate that Chinese CLIP can achieve the
state-of-the-art performance on MUGE, Flickr30K-CN, and COCO-CN in the setups
of zero-shot learning and finetuning, and it is able to achieve competitive
performance in zero-shot image classification based on the evaluation on the
ELEVATER benchmark (Li et al., 2022). We have released our codes, models, and
demos in https://github.com/OFA-Sys/Chinese-CLIP
| [
{
"version": "v1",
"created": "Wed, 2 Nov 2022 17:47:23 GMT"
},
{
"version": "v2",
"created": "Thu, 3 Nov 2022 13:21:44 GMT"
},
{
"version": "v3",
"created": "Tue, 23 May 2023 01:28:21 GMT"
}
] | 2023-05-24T00:00:00 | [
[
"Yang",
"An",
""
],
[
"Pan",
"Junshu",
""
],
[
"Lin",
"Junyang",
""
],
[
"Men",
"Rui",
""
],
[
"Zhang",
"Yichang",
""
],
[
"Zhou",
"Jingren",
""
],
[
"Zhou",
"Chang",
""
]
] | TITLE: Chinese CLIP: Contrastive Vision-Language Pretraining in Chinese
ABSTRACT: The tremendous success of CLIP (Radford et al., 2021) has promoted the
research and application of contrastive learning for vision-language
pretraining. In this work, we construct a large-scale dataset of image-text
pairs in Chinese, where most data are retrieved from publicly available
datasets, and we pretrain Chinese CLIP models on the new dataset. We develop 5
Chinese CLIP models of multiple sizes, spanning from 77 to 958 million
parameters. Furthermore, we propose a two-stage pretraining method, where the
model is first trained with the image encoder frozen and then trained with all
parameters being optimized, to achieve enhanced model performance. Our
comprehensive experiments demonstrate that Chinese CLIP can achieve the
state-of-the-art performance on MUGE, Flickr30K-CN, and COCO-CN in the setups
of zero-shot learning and finetuning, and it is able to achieve competitive
performance in zero-shot image classification based on the evaluation on the
ELEVATER benchmark (Li et al., 2022). We have released our codes, models, and
demos in https://github.com/OFA-Sys/Chinese-CLIP
| new_dataset | 0.717983 |
2211.02527 | Fran\c{c}ois G. Germain | Hao Yen, Fran\c{c}ois G. Germain, Gordon Wichern, Jonathan Le Roux | Cold Diffusion for Speech Enhancement | 5 pages, 1 figure, 1 table, 3 algorithms. To appear in ICASSP 2023.
With corrected references | null | null | null | eess.AS cs.SD | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Diffusion models have recently shown promising results for difficult
enhancement tasks such as the conditional and unconditional restoration of
natural images and audio signals. In this work, we explore the possibility of
leveraging a recently proposed advanced iterative diffusion model, namely cold
diffusion, to recover clean speech signals from noisy signals. The unique
mathematical properties of the sampling process from cold diffusion could be
utilized to restore high-quality samples from arbitrary degradations. Based on
these properties, we propose an improved training algorithm and objective to
help the model generalize better during the sampling process. We verify our
proposed framework by investigating two model architectures. Experimental
results on benchmark speech enhancement dataset VoiceBank-DEMAND demonstrate
the strong performance of the proposed approach compared to representative
discriminative models and diffusion-based enhancement models.
| [
{
"version": "v1",
"created": "Fri, 4 Nov 2022 15:41:01 GMT"
},
{
"version": "v2",
"created": "Tue, 28 Mar 2023 20:46:34 GMT"
},
{
"version": "v3",
"created": "Tue, 23 May 2023 17:56:02 GMT"
}
] | 2023-05-24T00:00:00 | [
[
"Yen",
"Hao",
""
],
[
"Germain",
"François G.",
""
],
[
"Wichern",
"Gordon",
""
],
[
"Roux",
"Jonathan Le",
""
]
] | TITLE: Cold Diffusion for Speech Enhancement
ABSTRACT: Diffusion models have recently shown promising results for difficult
enhancement tasks such as the conditional and unconditional restoration of
natural images and audio signals. In this work, we explore the possibility of
leveraging a recently proposed advanced iterative diffusion model, namely cold
diffusion, to recover clean speech signals from noisy signals. The unique
mathematical properties of the sampling process from cold diffusion could be
utilized to restore high-quality samples from arbitrary degradations. Based on
these properties, we propose an improved training algorithm and objective to
help the model generalize better during the sampling process. We verify our
proposed framework by investigating two model architectures. Experimental
results on benchmark speech enhancement dataset VoiceBank-DEMAND demonstrate
the strong performance of the proposed approach compared to representative
discriminative models and diffusion-based enhancement models.
| no_new_dataset | 0.710986 |
2211.04928 | Tassilo Klein | Tassilo Klein and Moin Nabi | miCSE: Mutual Information Contrastive Learning for Low-shot Sentence
Embeddings | Accepted to ACL 2023 | null | null | null | cs.CL cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | This paper presents miCSE, a mutual information-based contrastive learning
framework that significantly advances the state-of-the-art in few-shot sentence
embedding. The proposed approach imposes alignment between the attention
pattern of different views during contrastive learning. Learning sentence
embeddings with miCSE entails enforcing the structural consistency across
augmented views for every sentence, making contrastive self-supervised learning
more sample efficient. As a result, the proposed approach shows strong
performance in the few-shot learning domain. While it achieves superior results
compared to state-of-the-art methods on multiple benchmarks in few-shot
learning, it is comparable in the full-shot scenario. This study opens up
avenues for efficient self-supervised learning methods that are more robust
than current contrastive methods for sentence embedding.
| [
{
"version": "v1",
"created": "Wed, 9 Nov 2022 14:57:37 GMT"
},
{
"version": "v2",
"created": "Tue, 23 May 2023 08:35:15 GMT"
}
] | 2023-05-24T00:00:00 | [
[
"Klein",
"Tassilo",
""
],
[
"Nabi",
"Moin",
""
]
] | TITLE: miCSE: Mutual Information Contrastive Learning for Low-shot Sentence
Embeddings
ABSTRACT: This paper presents miCSE, a mutual information-based contrastive learning
framework that significantly advances the state-of-the-art in few-shot sentence
embedding. The proposed approach imposes alignment between the attention
pattern of different views during contrastive learning. Learning sentence
embeddings with miCSE entails enforcing the structural consistency across
augmented views for every sentence, making contrastive self-supervised learning
more sample efficient. As a result, the proposed approach shows strong
performance in the few-shot learning domain. While it achieves superior results
compared to state-of-the-art methods on multiple benchmarks in few-shot
learning, it is comparable in the full-shot scenario. This study opens up
avenues for efficient self-supervised learning methods that are more robust
than current contrastive methods for sentence embedding.
| no_new_dataset | 0.712201 |
2211.07615 | Sagar Gubbi Venkatesh | Sagar Gubbi Venkatesh, Partha Talukdar, Srini Narayanan | UGIF: UI Grounded Instruction Following | null | null | null | null | cs.CL | http://creativecommons.org/licenses/by/4.0/ | Smartphone users often find it difficult to navigate myriad menus to perform
common tasks such as "How to block calls from unknown numbers?". Currently,
help documents with step-by-step instructions are manually written to aid the
user. The user experience can be further enhanced by grounding the instructions
in the help document to the UI and overlaying a tutorial on the phone UI. To
build such tutorials, several natural language processing components including
retrieval, parsing, and grounding are necessary, but there isn't any relevant
dataset for such a task. Thus, we introduce UGIF-DataSet, a multi-lingual,
multi-modal UI grounded dataset for step-by-step task completion on the
smartphone containing 4,184 tasks across 8 languages. As an initial approach to
this problem, we propose retrieving the relevant instruction steps based on the
user's query and parsing the steps using Large Language Models (LLMs) to
generate macros that can be executed on-device. The instruction steps are often
available only in English, so the challenge includes cross-modal, cross-lingual
retrieval of English how-to pages from user queries in many languages and
mapping English instruction steps to UI in a potentially different language. We
compare the performance of different LLMs including PaLM and GPT-3 and find
that the end-to-end task completion rate is 48% for English UI but the
performance drops to 32% for other languages. We analyze the common failure
modes of existing models on this task and point out areas for improvement.
| [
{
"version": "v1",
"created": "Mon, 14 Nov 2022 18:36:19 GMT"
},
{
"version": "v2",
"created": "Tue, 23 May 2023 16:08:10 GMT"
}
] | 2023-05-24T00:00:00 | [
[
"Venkatesh",
"Sagar Gubbi",
""
],
[
"Talukdar",
"Partha",
""
],
[
"Narayanan",
"Srini",
""
]
] | TITLE: UGIF: UI Grounded Instruction Following
ABSTRACT: Smartphone users often find it difficult to navigate myriad menus to perform
common tasks such as "How to block calls from unknown numbers?". Currently,
help documents with step-by-step instructions are manually written to aid the
user. The user experience can be further enhanced by grounding the instructions
in the help document to the UI and overlaying a tutorial on the phone UI. To
build such tutorials, several natural language processing components including
retrieval, parsing, and grounding are necessary, but there isn't any relevant
dataset for such a task. Thus, we introduce UGIF-DataSet, a multi-lingual,
multi-modal UI grounded dataset for step-by-step task completion on the
smartphone containing 4,184 tasks across 8 languages. As an initial approach to
this problem, we propose retrieving the relevant instruction steps based on the
user's query and parsing the steps using Large Language Models (LLMs) to
generate macros that can be executed on-device. The instruction steps are often
available only in English, so the challenge includes cross-modal, cross-lingual
retrieval of English how-to pages from user queries in many languages and
mapping English instruction steps to UI in a potentially different language. We
compare the performance of different LLMs including PaLM and GPT-3 and find
that the end-to-end task completion rate is 48% for English UI but the
performance drops to 32% for other languages. We analyze the common failure
modes of existing models on this task and point out areas for improvement.
| new_dataset | 0.719168 |
2211.15510 | Nicolas Michael M\"uller | Nicolas M. M\"uller, Jochen Jacobs, Jennifer Williams, Konstantin
B\"ottinger | Localized Shortcut Removal | Accepted at XAI4CV @ CVPR2023 | null | null | null | cs.CV cs.LG eess.IV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Machine learning is a data-driven field, and the quality of the underlying
datasets plays a crucial role in learning success. However, high performance on
held-out test data does not necessarily indicate that a model generalizes or
learns anything meaningful. This is often due to the existence of machine
learning shortcuts - features in the data that are predictive but unrelated to
the problem at hand. To address this issue for datasets where the shortcuts are
smaller and more localized than true features, we propose a novel approach to
detect and remove them. We use an adversarially trained lens to detect and
eliminate highly predictive but semantically unconnected clues in images. In
our experiments on both synthetic and real-world data, we show that our
proposed approach reliably identifies and neutralizes such shortcuts without
causing degradation of model performance on clean data. We believe that our
approach can lead to more meaningful and generalizable machine learning models,
especially in scenarios where the quality of the underlying datasets is
crucial.
| [
{
"version": "v1",
"created": "Thu, 24 Nov 2022 13:05:33 GMT"
},
{
"version": "v2",
"created": "Tue, 23 May 2023 08:27:43 GMT"
}
] | 2023-05-24T00:00:00 | [
[
"Müller",
"Nicolas M.",
""
],
[
"Jacobs",
"Jochen",
""
],
[
"Williams",
"Jennifer",
""
],
[
"Böttinger",
"Konstantin",
""
]
] | TITLE: Localized Shortcut Removal
ABSTRACT: Machine learning is a data-driven field, and the quality of the underlying
datasets plays a crucial role in learning success. However, high performance on
held-out test data does not necessarily indicate that a model generalizes or
learns anything meaningful. This is often due to the existence of machine
learning shortcuts - features in the data that are predictive but unrelated to
the problem at hand. To address this issue for datasets where the shortcuts are
smaller and more localized than true features, we propose a novel approach to
detect and remove them. We use an adversarially trained lens to detect and
eliminate highly predictive but semantically unconnected clues in images. In
our experiments on both synthetic and real-world data, we show that our
proposed approach reliably identifies and neutralizes such shortcuts without
causing degradation of model performance on clean data. We believe that our
approach can lead to more meaningful and generalizable machine learning models,
especially in scenarios where the quality of the underlying datasets is
crucial.
| no_new_dataset | 0.712001 |
2211.15914 | Adithya Bhaskar | Adithya Bhaskar, Alexander R. Fabbri and Greg Durrett | Prompted Opinion Summarization with GPT-3.5 | Accepted to ACL (Findings) 2023 | null | null | null | cs.CL | http://creativecommons.org/licenses/by/4.0/ | Large language models have shown impressive performance across a wide variety
of tasks, including text summarization. In this paper, we show that this strong
performance extends to opinion summarization. We explore several pipeline
methods for applying GPT-3.5 to summarize a large collection of user reviews in
a prompted fashion. To handle arbitrarily large numbers of user reviews, we
explore recursive summarization as well as methods for selecting salient
content to summarize through supervised clustering or extraction. On two
datasets, an aspect-oriented summarization dataset of hotel reviews (SPACE) and
a generic summarization dataset of Amazon and Yelp reviews (FewSum), we show
that GPT-3.5 models achieve very strong performance in human evaluation. We
argue that standard evaluation metrics do not reflect this, and introduce three
new metrics targeting faithfulness, factuality, and genericity to contrast
these different methods.
| [
{
"version": "v1",
"created": "Tue, 29 Nov 2022 04:06:21 GMT"
},
{
"version": "v2",
"created": "Tue, 23 May 2023 06:19:47 GMT"
}
] | 2023-05-24T00:00:00 | [
[
"Bhaskar",
"Adithya",
""
],
[
"Fabbri",
"Alexander R.",
""
],
[
"Durrett",
"Greg",
""
]
] | TITLE: Prompted Opinion Summarization with GPT-3.5
ABSTRACT: Large language models have shown impressive performance across a wide variety
of tasks, including text summarization. In this paper, we show that this strong
performance extends to opinion summarization. We explore several pipeline
methods for applying GPT-3.5 to summarize a large collection of user reviews in
a prompted fashion. To handle arbitrarily large numbers of user reviews, we
explore recursive summarization as well as methods for selecting salient
content to summarize through supervised clustering or extraction. On two
datasets, an aspect-oriented summarization dataset of hotel reviews (SPACE) and
a generic summarization dataset of Amazon and Yelp reviews (FewSum), we show
that GPT-3.5 models achieve very strong performance in human evaluation. We
argue that standard evaluation metrics do not reflect this, and introduce three
new metrics targeting faithfulness, factuality, and genericity to contrast
these different methods.
| new_dataset | 0.692837 |
2212.01762 | Hsin-Ping Huang | Hsin-Ping Huang, Charles Herrmann, Junhwa Hur, Erika Lu, Kyle Sargent,
Austin Stone, Ming-Hsuan Yang, Deqing Sun | Self-supervised AutoFlow | null | null | null | null | cs.CV | http://creativecommons.org/licenses/by-nc-sa/4.0/ | Recently, AutoFlow has shown promising results on learning a training set for
optical flow, but requires ground truth labels in the target domain to compute
its search metric. Observing a strong correlation between the ground truth
search metric and self-supervised losses, we introduce self-supervised AutoFlow
to handle real-world videos without ground truth labels. Using self-supervised
loss as the search metric, our self-supervised AutoFlow performs on par with
AutoFlow on Sintel and KITTI where ground truth is available, and performs
better on the real-world DAVIS dataset. We further explore using
self-supervised AutoFlow in the (semi-)supervised setting and obtain
competitive results against the state of the art.
| [
{
"version": "v1",
"created": "Sun, 4 Dec 2022 08:09:12 GMT"
},
{
"version": "v2",
"created": "Thu, 8 Dec 2022 01:11:34 GMT"
},
{
"version": "v3",
"created": "Mon, 22 May 2023 21:14:51 GMT"
}
] | 2023-05-24T00:00:00 | [
[
"Huang",
"Hsin-Ping",
""
],
[
"Herrmann",
"Charles",
""
],
[
"Hur",
"Junhwa",
""
],
[
"Lu",
"Erika",
""
],
[
"Sargent",
"Kyle",
""
],
[
"Stone",
"Austin",
""
],
[
"Yang",
"Ming-Hsuan",
""
],
[
"Sun",
"Deqing",
""
]
] | TITLE: Self-supervised AutoFlow
ABSTRACT: Recently, AutoFlow has shown promising results on learning a training set for
optical flow, but requires ground truth labels in the target domain to compute
its search metric. Observing a strong correlation between the ground truth
search metric and self-supervised losses, we introduce self-supervised AutoFlow
to handle real-world videos without ground truth labels. Using self-supervised
loss as the search metric, our self-supervised AutoFlow performs on par with
AutoFlow on Sintel and KITTI where ground truth is available, and performs
better on the real-world DAVIS dataset. We further explore using
self-supervised AutoFlow in the (semi-)supervised setting and obtain
competitive results against the state of the art.
| no_new_dataset | 0.710007 |
2212.08192 | Martin P\"omsl | Akshatha Arodi, Martin P\"omsl, Kaheer Suleman, Adam Trischler,
Alexandra Olteanu, Jackie Chi Kit Cheung | The KITMUS Test: Evaluating Knowledge Integration from Multiple Sources
in Natural Language Understanding Systems | Accepted at ACL 2023. Code available at
https://github.com/mpoemsl/kitmus | null | null | null | cs.CL cs.LG | http://creativecommons.org/licenses/by-sa/4.0/ | Many state-of-the-art natural language understanding (NLU) models are based
on pretrained neural language models. These models often make inferences using
information from multiple sources. An important class of such inferences are
those that require both background knowledge, presumably contained in a model's
pretrained parameters, and instance-specific information that is supplied at
inference time. However, the integration and reasoning abilities of NLU models
in the presence of multiple knowledge sources have been largely understudied.
In this work, we propose a test suite of coreference resolution subtasks that
require reasoning over multiple facts. These subtasks differ in terms of which
knowledge sources contain the relevant facts. We also introduce subtasks where
knowledge is present only at inference time using fictional knowledge. We
evaluate state-of-the-art coreference resolution models on our dataset. Our
results indicate that several models struggle to reason on-the-fly over
knowledge observed both at pretrain time and at inference time. However, with
task-specific training, a subset of models demonstrates the ability to
integrate certain knowledge types from multiple sources. Still, even the best
performing models seem to have difficulties with reliably integrating knowledge
presented only at inference time.
| [
{
"version": "v1",
"created": "Thu, 15 Dec 2022 23:26:54 GMT"
},
{
"version": "v2",
"created": "Mon, 22 May 2023 21:17:52 GMT"
}
] | 2023-05-24T00:00:00 | [
[
"Arodi",
"Akshatha",
""
],
[
"Pömsl",
"Martin",
""
],
[
"Suleman",
"Kaheer",
""
],
[
"Trischler",
"Adam",
""
],
[
"Olteanu",
"Alexandra",
""
],
[
"Cheung",
"Jackie Chi Kit",
""
]
] | TITLE: The KITMUS Test: Evaluating Knowledge Integration from Multiple Sources
in Natural Language Understanding Systems
ABSTRACT: Many state-of-the-art natural language understanding (NLU) models are based
on pretrained neural language models. These models often make inferences using
information from multiple sources. An important class of such inferences are
those that require both background knowledge, presumably contained in a model's
pretrained parameters, and instance-specific information that is supplied at
inference time. However, the integration and reasoning abilities of NLU models
in the presence of multiple knowledge sources have been largely understudied.
In this work, we propose a test suite of coreference resolution subtasks that
require reasoning over multiple facts. These subtasks differ in terms of which
knowledge sources contain the relevant facts. We also introduce subtasks where
knowledge is present only at inference time using fictional knowledge. We
evaluate state-of-the-art coreference resolution models on our dataset. Our
results indicate that several models struggle to reason on-the-fly over
knowledge observed both at pretrain time and at inference time. However, with
task-specific training, a subset of models demonstrates the ability to
integrate certain knowledge types from multiple sources. Still, even the best
performing models seem to have difficulties with reliably integrating knowledge
presented only at inference time.
| new_dataset | 0.718746 |
2301.00427 | Han Huang | Han Huang, Leilei Sun, Bowen Du, Weifeng Lv | Conditional Diffusion Based on Discrete Graph Structures for Molecular
Graph Generation | Accepted by AAAI 2023 | null | null | null | cs.LG q-bio.BM | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Learning the underlying distribution of molecular graphs and generating
high-fidelity samples is a fundamental research problem in drug discovery and
material science. However, accurately modeling distribution and rapidly
generating novel molecular graphs remain crucial and challenging goals. To
accomplish these goals, we propose a novel Conditional Diffusion model based on
discrete Graph Structures (CDGS) for molecular graph generation. Specifically,
we construct a forward graph diffusion process on both graph structures and
inherent features through stochastic differential equations (SDE) and derive
discrete graph structures as the condition for reverse generative processes. We
present a specialized hybrid graph noise prediction model that extracts the
global context and the local node-edge dependency from intermediate graph
states. We further utilize ordinary differential equation (ODE) solvers for
efficient graph sampling, based on the semi-linear structure of the probability
flow ODE. Experiments on diverse datasets validate the effectiveness of our
framework. Particularly, the proposed method still generates high-quality
molecular graphs in a limited number of steps. Our code is provided in
https://github.com/GRAPH-0/CDGS.
| [
{
"version": "v1",
"created": "Sun, 1 Jan 2023 15:24:15 GMT"
},
{
"version": "v2",
"created": "Tue, 23 May 2023 09:41:57 GMT"
}
] | 2023-05-24T00:00:00 | [
[
"Huang",
"Han",
""
],
[
"Sun",
"Leilei",
""
],
[
"Du",
"Bowen",
""
],
[
"Lv",
"Weifeng",
""
]
] | TITLE: Conditional Diffusion Based on Discrete Graph Structures for Molecular
Graph Generation
ABSTRACT: Learning the underlying distribution of molecular graphs and generating
high-fidelity samples is a fundamental research problem in drug discovery and
material science. However, accurately modeling distribution and rapidly
generating novel molecular graphs remain crucial and challenging goals. To
accomplish these goals, we propose a novel Conditional Diffusion model based on
discrete Graph Structures (CDGS) for molecular graph generation. Specifically,
we construct a forward graph diffusion process on both graph structures and
inherent features through stochastic differential equations (SDE) and derive
discrete graph structures as the condition for reverse generative processes. We
present a specialized hybrid graph noise prediction model that extracts the
global context and the local node-edge dependency from intermediate graph
states. We further utilize ordinary differential equation (ODE) solvers for
efficient graph sampling, based on the semi-linear structure of the probability
flow ODE. Experiments on diverse datasets validate the effectiveness of our
framework. Particularly, the proposed method still generates high-quality
molecular graphs in a limited number of steps. Our code is provided in
https://github.com/GRAPH-0/CDGS.
| no_new_dataset | 0.709988 |
2302.06245 | Linwei Tao | Linwei Tao, Minjing Dong, Daochang Liu, Changming Sun, Chang Xu | Calibrating a Deep Neural Network with Its Predecessors | IJCAI 2023 Accept | null | null | null | cs.LG | http://creativecommons.org/licenses/by/4.0/ | Confidence calibration - the process to calibrate the output probability
distribution of neural networks - is essential for safety-critical applications
of such networks. Recent works verify the link between mis-calibration and
overfitting. However, early stopping, as a well-known technique to mitigate
overfitting, fails to calibrate networks. In this work, we study the limitions
of early stopping and comprehensively analyze the overfitting problem of a
network considering each individual block. We then propose a novel
regularization method, predecessor combination search (PCS), to improve
calibration by searching a combination of best-fitting block predecessors,
where block predecessors are the corresponding network blocks with weight
parameters from earlier training stages. PCS achieves the state-of-the-art
calibration performance on multiple datasets and architectures. In addition,
PCS improves model robustness under dataset distribution shift.
| [
{
"version": "v1",
"created": "Mon, 13 Feb 2023 10:33:55 GMT"
},
{
"version": "v2",
"created": "Tue, 23 May 2023 04:24:56 GMT"
}
] | 2023-05-24T00:00:00 | [
[
"Tao",
"Linwei",
""
],
[
"Dong",
"Minjing",
""
],
[
"Liu",
"Daochang",
""
],
[
"Sun",
"Changming",
""
],
[
"Xu",
"Chang",
""
]
] | TITLE: Calibrating a Deep Neural Network with Its Predecessors
ABSTRACT: Confidence calibration - the process to calibrate the output probability
distribution of neural networks - is essential for safety-critical applications
of such networks. Recent works verify the link between mis-calibration and
overfitting. However, early stopping, as a well-known technique to mitigate
overfitting, fails to calibrate networks. In this work, we study the limitions
of early stopping and comprehensively analyze the overfitting problem of a
network considering each individual block. We then propose a novel
regularization method, predecessor combination search (PCS), to improve
calibration by searching a combination of best-fitting block predecessors,
where block predecessors are the corresponding network blocks with weight
parameters from earlier training stages. PCS achieves the state-of-the-art
calibration performance on multiple datasets and architectures. In addition,
PCS improves model robustness under dataset distribution shift.
| no_new_dataset | 0.708824 |
2302.11327 | Seyedsaman Emami | Seyedsaman Emami and Gonzalo Mart\'inez-Mu\~noz | A Gradient Boosting Approach for Training Convolutional and Deep Neural
Networks | null | null | 10.1109/OJSP.2023.3279011 | null | cs.CV | http://creativecommons.org/licenses/by-nc-nd/4.0/ | Deep learning has revolutionized the computer vision and image classification
domains. In this context Convolutional Neural Networks (CNNs) based
architectures are the most widely applied models. In this article, we
introduced two procedures for training Convolutional Neural Networks (CNNs) and
Deep Neural Network based on Gradient Boosting (GB), namely GB-CNN and GB-DNN.
These models are trained to fit the gradient of the loss function or
pseudo-residuals of previous models. At each iteration, the proposed method
adds one dense layer to an exact copy of the previous deep NN model. The
weights of the dense layers trained on previous iterations are frozen to
prevent over-fitting, permitting the model to fit the new dense as well as to
fine-tune the convolutional layers (for GB-CNN) while still utilizing the
information already learned. Through extensive experimentation on different
2D-image classification and tabular datasets, the presented models show
superior performance in terms of classification accuracy with respect to
standard CNN and Deep-NN with the same architectures.
| [
{
"version": "v1",
"created": "Wed, 22 Feb 2023 12:17:32 GMT"
},
{
"version": "v2",
"created": "Thu, 23 Feb 2023 09:13:03 GMT"
}
] | 2023-05-24T00:00:00 | [
[
"Emami",
"Seyedsaman",
""
],
[
"Martínez-Muñoz",
"Gonzalo",
""
]
] | TITLE: A Gradient Boosting Approach for Training Convolutional and Deep Neural
Networks
ABSTRACT: Deep learning has revolutionized the computer vision and image classification
domains. In this context Convolutional Neural Networks (CNNs) based
architectures are the most widely applied models. In this article, we
introduced two procedures for training Convolutional Neural Networks (CNNs) and
Deep Neural Network based on Gradient Boosting (GB), namely GB-CNN and GB-DNN.
These models are trained to fit the gradient of the loss function or
pseudo-residuals of previous models. At each iteration, the proposed method
adds one dense layer to an exact copy of the previous deep NN model. The
weights of the dense layers trained on previous iterations are frozen to
prevent over-fitting, permitting the model to fit the new dense as well as to
fine-tune the convolutional layers (for GB-CNN) while still utilizing the
information already learned. Through extensive experimentation on different
2D-image classification and tabular datasets, the presented models show
superior performance in terms of classification accuracy with respect to
standard CNN and Deep-NN with the same architectures.
| no_new_dataset | 0.71262 |
2303.01904 | Junha Song | Junha Song, Jungsoo Lee, In So Kweon, Sungha Choi | EcoTTA: Memory-Efficient Continual Test-time Adaptation via
Self-distilled Regularization | Accepted to CVPR 2023, Project page:
https://sites.google.com/view/junha/ecotta | null | null | null | cs.CV | http://creativecommons.org/licenses/by/4.0/ | This paper presents a simple yet effective approach that improves continual
test-time adaptation (TTA) in a memory-efficient manner. TTA may primarily be
conducted on edge devices with limited memory, so reducing memory is crucial
but has been overlooked in previous TTA studies. In addition, long-term
adaptation often leads to catastrophic forgetting and error accumulation, which
hinders applying TTA in real-world deployments. Our approach consists of two
components to address these issues. First, we present lightweight meta networks
that can adapt the frozen original networks to the target domain. This novel
architecture minimizes memory consumption by decreasing the size of
intermediate activations required for backpropagation. Second, our novel
self-distilled regularization controls the output of the meta networks not to
deviate significantly from the output of the frozen original networks, thereby
preserving well-trained knowledge from the source domain. Without additional
memory, this regularization prevents error accumulation and catastrophic
forgetting, resulting in stable performance even in long-term test-time
adaptation. We demonstrate that our simple yet effective strategy outperforms
other state-of-the-art methods on various benchmarks for image classification
and semantic segmentation tasks. Notably, our proposed method with ResNet-50
and WideResNet-40 takes 86% and 80% less memory than the recent
state-of-the-art method, CoTTA.
| [
{
"version": "v1",
"created": "Fri, 3 Mar 2023 13:05:30 GMT"
},
{
"version": "v2",
"created": "Mon, 6 Mar 2023 03:51:39 GMT"
},
{
"version": "v3",
"created": "Mon, 13 Mar 2023 05:52:56 GMT"
},
{
"version": "v4",
"created": "Tue, 23 May 2023 05:33:02 GMT"
}
] | 2023-05-24T00:00:00 | [
[
"Song",
"Junha",
""
],
[
"Lee",
"Jungsoo",
""
],
[
"Kweon",
"In So",
""
],
[
"Choi",
"Sungha",
""
]
] | TITLE: EcoTTA: Memory-Efficient Continual Test-time Adaptation via
Self-distilled Regularization
ABSTRACT: This paper presents a simple yet effective approach that improves continual
test-time adaptation (TTA) in a memory-efficient manner. TTA may primarily be
conducted on edge devices with limited memory, so reducing memory is crucial
but has been overlooked in previous TTA studies. In addition, long-term
adaptation often leads to catastrophic forgetting and error accumulation, which
hinders applying TTA in real-world deployments. Our approach consists of two
components to address these issues. First, we present lightweight meta networks
that can adapt the frozen original networks to the target domain. This novel
architecture minimizes memory consumption by decreasing the size of
intermediate activations required for backpropagation. Second, our novel
self-distilled regularization controls the output of the meta networks not to
deviate significantly from the output of the frozen original networks, thereby
preserving well-trained knowledge from the source domain. Without additional
memory, this regularization prevents error accumulation and catastrophic
forgetting, resulting in stable performance even in long-term test-time
adaptation. We demonstrate that our simple yet effective strategy outperforms
other state-of-the-art methods on various benchmarks for image classification
and semantic segmentation tasks. Notably, our proposed method with ResNet-50
and WideResNet-40 takes 86% and 80% less memory than the recent
state-of-the-art method, CoTTA.
| no_new_dataset | 0.709636 |
2303.08958 | Talip Ucar | Talip Ucar | NESS: Node Embeddings from Static SubGraphs | 16 pages | null | null | null | cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We present a framework for learning Node Embeddings from Static Subgraphs
(NESS) using a graph autoencoder (GAE) in a transductive setting. NESS is based
on two key ideas: i) Partitioning the training graph to multiple static, sparse
subgraphs with non-overlapping edges using random edge split during data
pre-processing, ii) Aggregating the node representations learned from each
subgraph to obtain a joint representation of the graph at test time. Moreover,
we propose an optional contrastive learning approach in transductive setting.
We demonstrate that NESS gives a better node representation for link prediction
tasks compared to current autoencoding methods that use either the whole graph
or stochastic subgraphs. Our experiments also show that NESS improves the
performance of a wide range of graph encoders and achieves state-of-the-art
results for link prediction on multiple real-world datasets with edge homophily
ratio ranging from strong heterophily to strong homophily.
| [
{
"version": "v1",
"created": "Wed, 15 Mar 2023 22:14:28 GMT"
},
{
"version": "v2",
"created": "Tue, 23 May 2023 09:21:14 GMT"
}
] | 2023-05-24T00:00:00 | [
[
"Ucar",
"Talip",
""
]
] | TITLE: NESS: Node Embeddings from Static SubGraphs
ABSTRACT: We present a framework for learning Node Embeddings from Static Subgraphs
(NESS) using a graph autoencoder (GAE) in a transductive setting. NESS is based
on two key ideas: i) Partitioning the training graph to multiple static, sparse
subgraphs with non-overlapping edges using random edge split during data
pre-processing, ii) Aggregating the node representations learned from each
subgraph to obtain a joint representation of the graph at test time. Moreover,
we propose an optional contrastive learning approach in transductive setting.
We demonstrate that NESS gives a better node representation for link prediction
tasks compared to current autoencoding methods that use either the whole graph
or stochastic subgraphs. Our experiments also show that NESS improves the
performance of a wide range of graph encoders and achieves state-of-the-art
results for link prediction on multiple real-world datasets with edge homophily
ratio ranging from strong heterophily to strong homophily.
| no_new_dataset | 0.709221 |
2303.15086 | Davide Moltisanti | Davide Moltisanti, Frank Keller, Hakan Bilen, Laura Sevilla-Lara | Learning Action Changes by Measuring Verb-Adverb Textual Relationships | CVPR 23. Version 2 updates some results due to an errata (see code
repository for more details). Code and dataset available at
https://github.com/dmoltisanti/air-cvpr23 | null | null | null | cs.CV | http://creativecommons.org/licenses/by/4.0/ | The goal of this work is to understand the way actions are performed in
videos. That is, given a video, we aim to predict an adverb indicating a
modification applied to the action (e.g. cut "finely"). We cast this problem as
a regression task. We measure textual relationships between verbs and adverbs
to generate a regression target representing the action change we aim to learn.
We test our approach on a range of datasets and achieve state-of-the-art
results on both adverb prediction and antonym classification. Furthermore, we
outperform previous work when we lift two commonly assumed conditions: the
availability of action labels during testing and the pairing of adverbs as
antonyms. Existing datasets for adverb recognition are either noisy, which
makes learning difficult, or contain actions whose appearance is not influenced
by adverbs, which makes evaluation less reliable. To address this, we collect a
new high quality dataset: Adverbs in Recipes (AIR). We focus on instructional
recipes videos, curating a set of actions that exhibit meaningful visual
changes when performed differently. Videos in AIR are more tightly trimmed and
were manually reviewed by multiple annotators to ensure high labelling quality.
Results show that models learn better from AIR given its cleaner videos. At the
same time, adverb prediction on AIR is challenging, demonstrating that there is
considerable room for improvement.
| [
{
"version": "v1",
"created": "Mon, 27 Mar 2023 10:53:38 GMT"
},
{
"version": "v2",
"created": "Tue, 23 May 2023 12:53:13 GMT"
}
] | 2023-05-24T00:00:00 | [
[
"Moltisanti",
"Davide",
""
],
[
"Keller",
"Frank",
""
],
[
"Bilen",
"Hakan",
""
],
[
"Sevilla-Lara",
"Laura",
""
]
] | TITLE: Learning Action Changes by Measuring Verb-Adverb Textual Relationships
ABSTRACT: The goal of this work is to understand the way actions are performed in
videos. That is, given a video, we aim to predict an adverb indicating a
modification applied to the action (e.g. cut "finely"). We cast this problem as
a regression task. We measure textual relationships between verbs and adverbs
to generate a regression target representing the action change we aim to learn.
We test our approach on a range of datasets and achieve state-of-the-art
results on both adverb prediction and antonym classification. Furthermore, we
outperform previous work when we lift two commonly assumed conditions: the
availability of action labels during testing and the pairing of adverbs as
antonyms. Existing datasets for adverb recognition are either noisy, which
makes learning difficult, or contain actions whose appearance is not influenced
by adverbs, which makes evaluation less reliable. To address this, we collect a
new high quality dataset: Adverbs in Recipes (AIR). We focus on instructional
recipes videos, curating a set of actions that exhibit meaningful visual
changes when performed differently. Videos in AIR are more tightly trimmed and
were manually reviewed by multiple annotators to ensure high labelling quality.
Results show that models learn better from AIR given its cleaner videos. At the
same time, adverb prediction on AIR is challenging, demonstrating that there is
considerable room for improvement.
| new_dataset | 0.714423 |
2303.17870 | Jian Ma | Jian Ma, Mingjun Zhao, Chen Chen, Ruichen Wang, Di Niu, Haonan Lu,
Xiaodong Lin | GlyphDraw: Seamlessly Rendering Text with Intricate Spatial Structures
in Text-to-Image Generation | 24 pages, 5 figures | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Recent breakthroughs in the field of language-guided image generation have
yielded impressive achievements, enabling the creation of high-quality and
diverse images based on user instructions.Although the synthesis performance is
fascinating, one significant limitation of current image generation models is
their insufficient ability to generate text coherently within images,
particularly for complex glyph structures like Chinese characters. To address
this problem, we introduce GlyphDraw, a general learning framework aiming to
endow image generation models with the capacity to generate images coherently
embedded with text for any specific language.We first sophisticatedly design
the image-text dataset's construction strategy, then build our model
specifically on a diffusion-based image generator and carefully modify the
network structure to allow the model to learn drawing language characters with
the help of glyph and position information.Furthermore, we maintain the model's
open-domain image synthesis capability by preventing catastrophic forgetting by
using parameter-efficient fine-tuning techniques.Extensive qualitative and
quantitative experiments demonstrate that our method not only produces accurate
language characters as in prompts, but also seamlessly blends the generated
text into the background.Please refer to our
\href{https://1073521013.github.io/glyph-draw.github.io/}{project page}.
\end{abstract}
| [
{
"version": "v1",
"created": "Fri, 31 Mar 2023 08:06:33 GMT"
},
{
"version": "v2",
"created": "Tue, 23 May 2023 04:07:00 GMT"
}
] | 2023-05-24T00:00:00 | [
[
"Ma",
"Jian",
""
],
[
"Zhao",
"Mingjun",
""
],
[
"Chen",
"Chen",
""
],
[
"Wang",
"Ruichen",
""
],
[
"Niu",
"Di",
""
],
[
"Lu",
"Haonan",
""
],
[
"Lin",
"Xiaodong",
""
]
] | TITLE: GlyphDraw: Seamlessly Rendering Text with Intricate Spatial Structures
in Text-to-Image Generation
ABSTRACT: Recent breakthroughs in the field of language-guided image generation have
yielded impressive achievements, enabling the creation of high-quality and
diverse images based on user instructions.Although the synthesis performance is
fascinating, one significant limitation of current image generation models is
their insufficient ability to generate text coherently within images,
particularly for complex glyph structures like Chinese characters. To address
this problem, we introduce GlyphDraw, a general learning framework aiming to
endow image generation models with the capacity to generate images coherently
embedded with text for any specific language.We first sophisticatedly design
the image-text dataset's construction strategy, then build our model
specifically on a diffusion-based image generator and carefully modify the
network structure to allow the model to learn drawing language characters with
the help of glyph and position information.Furthermore, we maintain the model's
open-domain image synthesis capability by preventing catastrophic forgetting by
using parameter-efficient fine-tuning techniques.Extensive qualitative and
quantitative experiments demonstrate that our method not only produces accurate
language characters as in prompts, but also seamlessly blends the generated
text into the background.Please refer to our
\href{https://1073521013.github.io/glyph-draw.github.io/}{project page}.
\end{abstract}
| no_new_dataset | 0.687263 |
2304.00354 | Bin Liu | Chenyang Zhao, Zihao Zhou, Bin Liu | On Context Distribution Shift in Task Representation Learning for
Offline Meta RL | This paper has been accepted by 19th Inter. Conf. on Intelligent
Computing (ICIC 2023) | null | null | null | cs.LG cs.AI | http://creativecommons.org/licenses/by-nc-nd/4.0/ | Offline Meta Reinforcement Learning (OMRL) aims to learn transferable
knowledge from offline datasets to enhance the learning process for new target
tasks. Context-based Reinforcement Learning (RL) adopts a context encoder to
expediently adapt the agent to new tasks by inferring the task representation,
and then adjusting the policy based on this inferred representation. In this
work, we focus on context-based OMRL, specifically on the challenge of learning
task representation for OMRL. We conduct experiments that demonstrate that the
context encoder trained on offline datasets might encounter distribution shift
between the contexts used for training and testing. To overcome this problem,
we present a hard-sampling-based strategy to train a robust task context
encoder. Our experimental findings on diverse continuous control tasks reveal
that utilizing our approach yields more robust task representations and better
testing performance in terms of accumulated returns compared to baseline
methods. Our code is available at https://github.com/ZJLAB-AMMI/HS-OMRL.
| [
{
"version": "v1",
"created": "Sat, 1 Apr 2023 16:21:55 GMT"
},
{
"version": "v2",
"created": "Tue, 23 May 2023 13:14:57 GMT"
}
] | 2023-05-24T00:00:00 | [
[
"Zhao",
"Chenyang",
""
],
[
"Zhou",
"Zihao",
""
],
[
"Liu",
"Bin",
""
]
] | TITLE: On Context Distribution Shift in Task Representation Learning for
Offline Meta RL
ABSTRACT: Offline Meta Reinforcement Learning (OMRL) aims to learn transferable
knowledge from offline datasets to enhance the learning process for new target
tasks. Context-based Reinforcement Learning (RL) adopts a context encoder to
expediently adapt the agent to new tasks by inferring the task representation,
and then adjusting the policy based on this inferred representation. In this
work, we focus on context-based OMRL, specifically on the challenge of learning
task representation for OMRL. We conduct experiments that demonstrate that the
context encoder trained on offline datasets might encounter distribution shift
between the contexts used for training and testing. To overcome this problem,
we present a hard-sampling-based strategy to train a robust task context
encoder. Our experimental findings on diverse continuous control tasks reveal
that utilizing our approach yields more robust task representations and better
testing performance in terms of accumulated returns compared to baseline
methods. Our code is available at https://github.com/ZJLAB-AMMI/HS-OMRL.
| no_new_dataset | 0.708584 |
2304.01236 | Anass Bairouk | Anass Bairouk, Marc Chaumont, Dominique Fouchez, Jerome Paquet,
Fr\'ed\'eric Comby, Julian Bautista | Astronomical image time series classification using CONVolutional
attENTION (ConvEntion) | null | A&A 673, A141 (2023) | 10.1051/0004-6361/202244657 | null | astro-ph.IM cs.LG | http://creativecommons.org/licenses/by-nc-sa/4.0/ | Aims. The treatment of astronomical image time series has won increasing
attention in recent years. Indeed, numerous surveys following up on transient
objects are in progress or under construction, such as the Vera Rubin
Observatory Legacy Survey for Space and Time (LSST), which is poised to produce
huge amounts of these time series. The associated scientific topics are
extensive, ranging from the study of objects in our galaxy to the observation
of the most distant supernovae for measuring the expansion of the universe.
With such a large amount of data available, the need for robust automatic tools
to detect and classify celestial objects is growing steadily. Methods. This
study is based on the assumption that astronomical images contain more
information than light curves. In this paper, we propose a novel approach based
on deep learning for classifying different types of space objects directly
using images. We named our approach ConvEntion, which stands for CONVolutional
attENTION. It is based on convolutions and transformers, which are new
approaches for the treatment of astronomical image time series. Our solution
integrates spatio-temporal features and can be applied to various types of
image datasets with any number of bands. Results. In this work, we solved
various problems the datasets tend to suffer from and we present new results
for classifications using astronomical image time series with an increase in
accuracy of 13%, compared to state-of-the-art approaches that use image time
series, and a 12% increase, compared to approaches that use light curves.
| [
{
"version": "v1",
"created": "Mon, 3 Apr 2023 08:48:44 GMT"
}
] | 2023-05-24T00:00:00 | [
[
"Bairouk",
"Anass",
""
],
[
"Chaumont",
"Marc",
""
],
[
"Fouchez",
"Dominique",
""
],
[
"Paquet",
"Jerome",
""
],
[
"Comby",
"Frédéric",
""
],
[
"Bautista",
"Julian",
""
]
] | TITLE: Astronomical image time series classification using CONVolutional
attENTION (ConvEntion)
ABSTRACT: Aims. The treatment of astronomical image time series has won increasing
attention in recent years. Indeed, numerous surveys following up on transient
objects are in progress or under construction, such as the Vera Rubin
Observatory Legacy Survey for Space and Time (LSST), which is poised to produce
huge amounts of these time series. The associated scientific topics are
extensive, ranging from the study of objects in our galaxy to the observation
of the most distant supernovae for measuring the expansion of the universe.
With such a large amount of data available, the need for robust automatic tools
to detect and classify celestial objects is growing steadily. Methods. This
study is based on the assumption that astronomical images contain more
information than light curves. In this paper, we propose a novel approach based
on deep learning for classifying different types of space objects directly
using images. We named our approach ConvEntion, which stands for CONVolutional
attENTION. It is based on convolutions and transformers, which are new
approaches for the treatment of astronomical image time series. Our solution
integrates spatio-temporal features and can be applied to various types of
image datasets with any number of bands. Results. In this work, we solved
various problems the datasets tend to suffer from and we present new results
for classifications using astronomical image time series with an increase in
accuracy of 13%, compared to state-of-the-art approaches that use image time
series, and a 12% increase, compared to approaches that use light curves.
| no_new_dataset | 0.712057 |
2304.03245 | Marzena Karpinska | Marzena Karpinska and Mohit Iyyer | Large language models effectively leverage document-level context for
literary translation, but critical errors persist | preprint (31 pages) | null | null | null | cs.CL | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Large language models (LLMs) are competitive with the state of the art on a
wide range of sentence-level translation datasets. However, their ability to
translate paragraphs and documents remains unexplored because evaluation in
these settings is costly and difficult. We show through a rigorous human
evaluation that asking the Gpt-3.5 (text-davinci-003) LLM to translate an
entire literary paragraph (e.g., from a novel) at once results in
higher-quality translations than standard sentence-by-sentence translation
across 18 linguistically-diverse language pairs (e.g., translating into and out
of Japanese, Polish, and English). Our evaluation, which took approximately 350
hours of effort for annotation and analysis, is conducted by hiring translators
fluent in both the source and target language and asking them to provide both
span-level error annotations as well as preference judgments of which system's
translations are better. We observe that discourse-level LLM translators commit
fewer mistranslations, grammar errors, and stylistic inconsistencies than
sentence-level approaches. With that said, critical errors still abound,
including occasional content omissions, and a human translator's intervention
remains necessary to ensure that the author's voice remains intact. We publicly
release our dataset and error annotations to spur future research on evaluation
of document-level literary translation.
| [
{
"version": "v1",
"created": "Thu, 6 Apr 2023 17:27:45 GMT"
},
{
"version": "v2",
"created": "Fri, 7 Apr 2023 12:52:03 GMT"
},
{
"version": "v3",
"created": "Mon, 22 May 2023 22:00:51 GMT"
}
] | 2023-05-24T00:00:00 | [
[
"Karpinska",
"Marzena",
""
],
[
"Iyyer",
"Mohit",
""
]
] | TITLE: Large language models effectively leverage document-level context for
literary translation, but critical errors persist
ABSTRACT: Large language models (LLMs) are competitive with the state of the art on a
wide range of sentence-level translation datasets. However, their ability to
translate paragraphs and documents remains unexplored because evaluation in
these settings is costly and difficult. We show through a rigorous human
evaluation that asking the Gpt-3.5 (text-davinci-003) LLM to translate an
entire literary paragraph (e.g., from a novel) at once results in
higher-quality translations than standard sentence-by-sentence translation
across 18 linguistically-diverse language pairs (e.g., translating into and out
of Japanese, Polish, and English). Our evaluation, which took approximately 350
hours of effort for annotation and analysis, is conducted by hiring translators
fluent in both the source and target language and asking them to provide both
span-level error annotations as well as preference judgments of which system's
translations are better. We observe that discourse-level LLM translators commit
fewer mistranslations, grammar errors, and stylistic inconsistencies than
sentence-level approaches. With that said, critical errors still abound,
including occasional content omissions, and a human translator's intervention
remains necessary to ensure that the author's voice remains intact. We publicly
release our dataset and error annotations to spur future research on evaluation
of document-level literary translation.
| new_dataset | 0.714454 |
2305.02073 | Xiaoyang Chen | Xiaoyang Chen, Yanjiang Liu, Ben He, Le Sun, Yingfei Sun | Understanding Differential Search Index for Text Retrieval | Accepted to Findings of ACL 2023 | null | null | null | cs.IR | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | The Differentiable Search Index (DSI) is a novel information retrieval (IR)
framework that utilizes a differentiable function to generate a sorted list of
document identifiers in response to a given query. However, due to the
black-box nature of the end-to-end neural architecture, it remains to be
understood to what extent DSI possesses the basic indexing and retrieval
abilities. To mitigate this gap, in this study, we define and examine three
important abilities that a functioning IR framework should possess, namely,
exclusivity, completeness, and relevance ordering. Our analytical
experimentation shows that while DSI demonstrates proficiency in memorizing the
unidirectional mapping from pseudo queries to document identifiers, it falls
short in distinguishing relevant documents from random ones, thereby negatively
impacting its retrieval effectiveness. To address this issue, we propose a
multi-task distillation approach to enhance the retrieval quality without
altering the structure of the model and successfully endow it with improved
indexing abilities. Through experiments conducted on various datasets, we
demonstrate that our proposed method outperforms previous DSI baselines.
| [
{
"version": "v1",
"created": "Wed, 3 May 2023 12:22:39 GMT"
},
{
"version": "v2",
"created": "Tue, 23 May 2023 13:07:16 GMT"
}
] | 2023-05-24T00:00:00 | [
[
"Chen",
"Xiaoyang",
""
],
[
"Liu",
"Yanjiang",
""
],
[
"He",
"Ben",
""
],
[
"Sun",
"Le",
""
],
[
"Sun",
"Yingfei",
""
]
] | TITLE: Understanding Differential Search Index for Text Retrieval
ABSTRACT: The Differentiable Search Index (DSI) is a novel information retrieval (IR)
framework that utilizes a differentiable function to generate a sorted list of
document identifiers in response to a given query. However, due to the
black-box nature of the end-to-end neural architecture, it remains to be
understood to what extent DSI possesses the basic indexing and retrieval
abilities. To mitigate this gap, in this study, we define and examine three
important abilities that a functioning IR framework should possess, namely,
exclusivity, completeness, and relevance ordering. Our analytical
experimentation shows that while DSI demonstrates proficiency in memorizing the
unidirectional mapping from pseudo queries to document identifiers, it falls
short in distinguishing relevant documents from random ones, thereby negatively
impacting its retrieval effectiveness. To address this issue, we propose a
multi-task distillation approach to enhance the retrieval quality without
altering the structure of the model and successfully endow it with improved
indexing abilities. Through experiments conducted on various datasets, we
demonstrate that our proposed method outperforms previous DSI baselines.
| no_new_dataset | 0.70695 |
2305.02783 | Saurabh Pujar | Saurabh Pujar, Luca Buratti, Xiaojie Guo, Nicolas Dupuis, Burn Lewis,
Sahil Suneja, Atin Sood, Ganesh Nalawade, Matthew Jones, Alessandro Morari,
Ruchir Puri | Automated Code generation for Information Technology Tasks in YAML
through Large Language Models | null | null | null | null | cs.SE cs.AI cs.CL cs.PL | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | The recent improvement in code generation capabilities due to the use of
large language models has mainly benefited general purpose programming
languages. Domain specific languages, such as the ones used for IT Automation,
have received far less attention, despite involving many active developers and
being an essential component of modern cloud platforms. This work focuses on
the generation of Ansible-YAML, a widely used markup language for IT
Automation. We present Ansible Wisdom, a natural-language to Ansible-YAML code
generation tool, aimed at improving IT automation productivity. Ansible Wisdom
is a transformer-based model, extended by training with a new dataset
containing Ansible-YAML. We also develop two novel performance metrics for YAML
and Ansible to capture the specific characteristics of this domain. Results
show that Ansible Wisdom can accurately generate Ansible script from natural
language prompts with performance comparable or better than existing state of
the art code generation models. In few-shot settings we asses the impact of
training with Ansible, YAML data and compare with different baselines including
Codex-Davinci-002. We also show that after finetuning, our Ansible specific
model (BLEU: 66.67) can outperform a much larger Codex-Davinci-002 (BLEU: 50.4)
model, which was evaluated in few shot settings.
| [
{
"version": "v1",
"created": "Tue, 2 May 2023 21:01:01 GMT"
},
{
"version": "v2",
"created": "Fri, 5 May 2023 00:42:03 GMT"
},
{
"version": "v3",
"created": "Tue, 16 May 2023 10:55:59 GMT"
},
{
"version": "v4",
"created": "Tue, 23 May 2023 15:45:42 GMT"
}
] | 2023-05-24T00:00:00 | [
[
"Pujar",
"Saurabh",
""
],
[
"Buratti",
"Luca",
""
],
[
"Guo",
"Xiaojie",
""
],
[
"Dupuis",
"Nicolas",
""
],
[
"Lewis",
"Burn",
""
],
[
"Suneja",
"Sahil",
""
],
[
"Sood",
"Atin",
""
],
[
"Nalawade",
"Ganesh",
""
],
[
"Jones",
"Matthew",
""
],
[
"Morari",
"Alessandro",
""
],
[
"Puri",
"Ruchir",
""
]
] | TITLE: Automated Code generation for Information Technology Tasks in YAML
through Large Language Models
ABSTRACT: The recent improvement in code generation capabilities due to the use of
large language models has mainly benefited general purpose programming
languages. Domain specific languages, such as the ones used for IT Automation,
have received far less attention, despite involving many active developers and
being an essential component of modern cloud platforms. This work focuses on
the generation of Ansible-YAML, a widely used markup language for IT
Automation. We present Ansible Wisdom, a natural-language to Ansible-YAML code
generation tool, aimed at improving IT automation productivity. Ansible Wisdom
is a transformer-based model, extended by training with a new dataset
containing Ansible-YAML. We also develop two novel performance metrics for YAML
and Ansible to capture the specific characteristics of this domain. Results
show that Ansible Wisdom can accurately generate Ansible script from natural
language prompts with performance comparable or better than existing state of
the art code generation models. In few-shot settings we asses the impact of
training with Ansible, YAML data and compare with different baselines including
Codex-Davinci-002. We also show that after finetuning, our Ansible specific
model (BLEU: 66.67) can outperform a much larger Codex-Davinci-002 (BLEU: 50.4)
model, which was evaluated in few shot settings.
| new_dataset | 0.715039 |
2305.02981 | Sergej Chicherin | Sergej Chicherin, Karen Efremyan | Adversarially-Guided Portrait Matting | null | null | null | null | cs.CV | http://creativecommons.org/licenses/by-sa/4.0/ | We present a method for generating alpha mattes using a limited data source.
We pretrain a novel transformerbased model (StyleMatte) on portrait datasets.
We utilize this model to provide image-mask pairs for the StyleGAN3-based
network (StyleMatteGAN). This network is trained unsupervisedly and generates
previously unseen imagemask training pairs that are fed back to StyleMatte. We
demonstrate that the performance of the matte pulling network improves during
this cycle and obtains top results on the human portraits and state-of-the-art
metrics on animals dataset. Furthermore, StyleMatteGAN provides
high-resolution, privacy-preserving portraits with alpha mattes, making it
suitable for various image composition tasks. Our code is available at
https://github.com/chroneus/stylematte
| [
{
"version": "v1",
"created": "Thu, 4 May 2023 16:45:04 GMT"
},
{
"version": "v2",
"created": "Tue, 23 May 2023 11:50:01 GMT"
}
] | 2023-05-24T00:00:00 | [
[
"Chicherin",
"Sergej",
""
],
[
"Efremyan",
"Karen",
""
]
] | TITLE: Adversarially-Guided Portrait Matting
ABSTRACT: We present a method for generating alpha mattes using a limited data source.
We pretrain a novel transformerbased model (StyleMatte) on portrait datasets.
We utilize this model to provide image-mask pairs for the StyleGAN3-based
network (StyleMatteGAN). This network is trained unsupervisedly and generates
previously unseen imagemask training pairs that are fed back to StyleMatte. We
demonstrate that the performance of the matte pulling network improves during
this cycle and obtains top results on the human portraits and state-of-the-art
metrics on animals dataset. Furthermore, StyleMatteGAN provides
high-resolution, privacy-preserving portraits with alpha mattes, making it
suitable for various image composition tasks. Our code is available at
https://github.com/chroneus/stylematte
| no_new_dataset | 0.710785 |
2305.03983 | Yang Wu | Yang Wu, Zhibin Liu, Hefeng Wu, Liang Lin | Multi-object Video Generation from Single Frame Layouts | 6 pages limit | null | null | null | cs.CV | http://creativecommons.org/licenses/by-sa/4.0/ | In this paper, we study video synthesis with emphasis on simplifying the
generation conditions. Most existing video synthesis models or datasets are
designed to address complex motions of a single object, lacking the ability of
comprehensively understanding the spatio-temporal relationships among multiple
objects. Besides, current methods are usually conditioned on intricate
annotations (e.g. video segmentations) to generate new videos, being
fundamentally less practical. These motivate us to generate multi-object videos
conditioning exclusively on object layouts from a single frame. To solve above
challenges and inspired by recent research on image generation from layouts, we
have proposed a novel video generative framework capable of synthesizing global
scenes with local objects, via implicit neural representations and layout
motion self-inference. Our framework is a non-trivial adaptation from image
generation methods, and is new to this field. In addition, our model has been
evaluated on two widely-used video recognition benchmarks, demonstrating
effectiveness compared to the baseline model.
| [
{
"version": "v1",
"created": "Sat, 6 May 2023 09:07:01 GMT"
},
{
"version": "v2",
"created": "Tue, 23 May 2023 15:52:48 GMT"
}
] | 2023-05-24T00:00:00 | [
[
"Wu",
"Yang",
""
],
[
"Liu",
"Zhibin",
""
],
[
"Wu",
"Hefeng",
""
],
[
"Lin",
"Liang",
""
]
] | TITLE: Multi-object Video Generation from Single Frame Layouts
ABSTRACT: In this paper, we study video synthesis with emphasis on simplifying the
generation conditions. Most existing video synthesis models or datasets are
designed to address complex motions of a single object, lacking the ability of
comprehensively understanding the spatio-temporal relationships among multiple
objects. Besides, current methods are usually conditioned on intricate
annotations (e.g. video segmentations) to generate new videos, being
fundamentally less practical. These motivate us to generate multi-object videos
conditioning exclusively on object layouts from a single frame. To solve above
challenges and inspired by recent research on image generation from layouts, we
have proposed a novel video generative framework capable of synthesizing global
scenes with local objects, via implicit neural representations and layout
motion self-inference. Our framework is a non-trivial adaptation from image
generation methods, and is new to this field. In addition, our model has been
evaluated on two widely-used video recognition benchmarks, demonstrating
effectiveness compared to the baseline model.
| no_new_dataset | 0.710879 |
2305.04532 | Tuomo Sipola | Teemu Niskanen, Tuomo Sipola, Olli V\"a\"an\"anen | Latest Trends in Artificial Intelligence Technology: A Scoping Review | null | null | null | null | cs.LG cs.AI cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Artificial intelligence is more ubiquitous in multiple domains. Smartphones,
social media platforms, search engines, and autonomous vehicles are just a few
examples of applications that utilize artificial intelligence technologies to
enhance their performance. This study carries out a scoping review of the
current state-of-the-art artificial intelligence technologies following the
PRISMA framework. The goal was to find the most advanced technologies used in
different domains of artificial intelligence technology research. Three
recognized journals were used from artificial intelligence and machine learning
domain: Journal of Artificial Intelligence Research, Journal of Machine
Learning Research, and Machine Learning, and articles published in 2022 were
observed. Certain qualifications were laid for the technological solutions: the
technology must be tested against comparable solutions, commonly approved or
otherwise well justified datasets must be used while applying, and results must
show improvements against comparable solutions. One of the most important parts
of the technology development appeared to be how to process and exploit the
data gathered from multiple sources. The data can be highly unstructured and
the technological solution should be able to utilize the data with minimum
manual work from humans. The results of this review indicate that creating
labeled datasets is very laborious, and solutions exploiting unsupervised or
semi-supervised learning technologies are more and more researched. The
learning algorithms should be able to be updated efficiently, and predictions
should be interpretable. Using artificial intelligence technologies in
real-world applications, safety and explainable predictions are mandatory to
consider before mass adoption can occur.
| [
{
"version": "v1",
"created": "Mon, 8 May 2023 08:06:16 GMT"
},
{
"version": "v2",
"created": "Tue, 23 May 2023 13:32:54 GMT"
}
] | 2023-05-24T00:00:00 | [
[
"Niskanen",
"Teemu",
""
],
[
"Sipola",
"Tuomo",
""
],
[
"Väänänen",
"Olli",
""
]
] | TITLE: Latest Trends in Artificial Intelligence Technology: A Scoping Review
ABSTRACT: Artificial intelligence is more ubiquitous in multiple domains. Smartphones,
social media platforms, search engines, and autonomous vehicles are just a few
examples of applications that utilize artificial intelligence technologies to
enhance their performance. This study carries out a scoping review of the
current state-of-the-art artificial intelligence technologies following the
PRISMA framework. The goal was to find the most advanced technologies used in
different domains of artificial intelligence technology research. Three
recognized journals were used from artificial intelligence and machine learning
domain: Journal of Artificial Intelligence Research, Journal of Machine
Learning Research, and Machine Learning, and articles published in 2022 were
observed. Certain qualifications were laid for the technological solutions: the
technology must be tested against comparable solutions, commonly approved or
otherwise well justified datasets must be used while applying, and results must
show improvements against comparable solutions. One of the most important parts
of the technology development appeared to be how to process and exploit the
data gathered from multiple sources. The data can be highly unstructured and
the technological solution should be able to utilize the data with minimum
manual work from humans. The results of this review indicate that creating
labeled datasets is very laborious, and solutions exploiting unsupervised or
semi-supervised learning technologies are more and more researched. The
learning algorithms should be able to be updated efficiently, and predictions
should be interpretable. Using artificial intelligence technologies in
real-world applications, safety and explainable predictions are mandatory to
consider before mass adoption can occur.
| no_new_dataset | 0.70866 |
2305.05988 | Pedro J. Martinez-Ferrer | Pedro J. Martinez-Ferrer, Tufan Arslan, Vicen\c{c} Beltran | Improving the performance of classical linear algebra iterative methods
via hybrid parallelism | 33 pages, 6 figures, accepted manuscript in Journal of Parallel and
Distributed Computing | Journal of Parallel and Distributed Computing, 179, (2023), 104711 | 10.1016/j.jpdc.2023.04.012 | null | cs.DC cs.DS cs.PF | http://creativecommons.org/licenses/by-nc-nd/4.0/ | We propose fork-join and task-based hybrid implementations of four classical
linear algebra iterative methods (Jacobi, Gauss-Seidel, conjugate gradient and
biconjugate gradient stabilised) as well as variations of them. Algorithms are
duly documented and the corresponding source code is made publicly available
for reproducibility. Both weak and strong scalability benchmarks are conducted
to statistically analyse their relative efficiencies.
The weak scalability results assert the superiority of a task-based hybrid
parallelisation over MPI-only and fork-join hybrid implementations. Indeed, the
task-based model is able to achieve speedups of up to 25% larger than its
MPI-only counterpart depending on the numerical method and the computational
resources used. For strong scalability scenarios, hybrid methods based on tasks
remain more efficient with moderate computational resources where data locality
does not play an important role. Fork-join hybridisation often yields mixed
results and hence does not present a competitive advantage over a much simpler
MPI approach.
| [
{
"version": "v1",
"created": "Wed, 10 May 2023 08:54:00 GMT"
},
{
"version": "v2",
"created": "Tue, 23 May 2023 09:46:14 GMT"
}
] | 2023-05-24T00:00:00 | [
[
"Martinez-Ferrer",
"Pedro J.",
""
],
[
"Arslan",
"Tufan",
""
],
[
"Beltran",
"Vicenç",
""
]
] | TITLE: Improving the performance of classical linear algebra iterative methods
via hybrid parallelism
ABSTRACT: We propose fork-join and task-based hybrid implementations of four classical
linear algebra iterative methods (Jacobi, Gauss-Seidel, conjugate gradient and
biconjugate gradient stabilised) as well as variations of them. Algorithms are
duly documented and the corresponding source code is made publicly available
for reproducibility. Both weak and strong scalability benchmarks are conducted
to statistically analyse their relative efficiencies.
The weak scalability results assert the superiority of a task-based hybrid
parallelisation over MPI-only and fork-join hybrid implementations. Indeed, the
task-based model is able to achieve speedups of up to 25% larger than its
MPI-only counterpart depending on the numerical method and the computational
resources used. For strong scalability scenarios, hybrid methods based on tasks
remain more efficient with moderate computational resources where data locality
does not play an important role. Fork-join hybridisation often yields mixed
results and hence does not present a competitive advantage over a much simpler
MPI approach.
| no_new_dataset | 0.708244 |
2305.06849 | Yujia Qin | Yujia Qin, Zihan Cai, Dian Jin, Lan Yan, Shihao Liang, Kunlun Zhu,
Yankai Lin, Xu Han, Ning Ding, Huadong Wang, Ruobing Xie, Fanchao Qi, Zhiyuan
Liu, Maosong Sun, and Jie Zhou | WebCPM: Interactive Web Search for Chinese Long-form Question Answering | ACL 2023, main conference | null | null | null | cs.CL cs.AI cs.IR | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Long-form question answering (LFQA) aims at answering complex, open-ended
questions with detailed, paragraph-length responses. The de facto paradigm of
LFQA necessitates two procedures: information retrieval, which searches for
relevant supporting facts, and information synthesis, which integrates these
facts into a coherent answer. In this paper, we introduce WebCPM, the first
Chinese LFQA dataset. One unique feature of WebCPM is that its information
retrieval is based on interactive web search, which engages with a search
engine in real time. Following WebGPT, we develop a web search interface. We
recruit annotators to search for relevant information using our interface and
then answer questions. Meanwhile, the web search behaviors of our annotators
would be recorded. In total, we collect 5,500 high-quality question-answer
pairs, together with 14,315 supporting facts and 121,330 web search actions. We
fine-tune pre-trained language models to imitate human behaviors for web search
and to generate answers based on the collected facts. Our LFQA pipeline, built
on these fine-tuned models, generates answers that are no worse than
human-written ones in 32.5% and 47.5% of the cases on our dataset and DuReader,
respectively.
| [
{
"version": "v1",
"created": "Thu, 11 May 2023 14:47:29 GMT"
},
{
"version": "v2",
"created": "Tue, 23 May 2023 13:15:10 GMT"
}
] | 2023-05-24T00:00:00 | [
[
"Qin",
"Yujia",
""
],
[
"Cai",
"Zihan",
""
],
[
"Jin",
"Dian",
""
],
[
"Yan",
"Lan",
""
],
[
"Liang",
"Shihao",
""
],
[
"Zhu",
"Kunlun",
""
],
[
"Lin",
"Yankai",
""
],
[
"Han",
"Xu",
""
],
[
"Ding",
"Ning",
""
],
[
"Wang",
"Huadong",
""
],
[
"Xie",
"Ruobing",
""
],
[
"Qi",
"Fanchao",
""
],
[
"Liu",
"Zhiyuan",
""
],
[
"Sun",
"Maosong",
""
],
[
"Zhou",
"Jie",
""
]
] | TITLE: WebCPM: Interactive Web Search for Chinese Long-form Question Answering
ABSTRACT: Long-form question answering (LFQA) aims at answering complex, open-ended
questions with detailed, paragraph-length responses. The de facto paradigm of
LFQA necessitates two procedures: information retrieval, which searches for
relevant supporting facts, and information synthesis, which integrates these
facts into a coherent answer. In this paper, we introduce WebCPM, the first
Chinese LFQA dataset. One unique feature of WebCPM is that its information
retrieval is based on interactive web search, which engages with a search
engine in real time. Following WebGPT, we develop a web search interface. We
recruit annotators to search for relevant information using our interface and
then answer questions. Meanwhile, the web search behaviors of our annotators
would be recorded. In total, we collect 5,500 high-quality question-answer
pairs, together with 14,315 supporting facts and 121,330 web search actions. We
fine-tune pre-trained language models to imitate human behaviors for web search
and to generate answers based on the collected facts. Our LFQA pipeline, built
on these fine-tuned models, generates answers that are no worse than
human-written ones in 32.5% and 47.5% of the cases on our dataset and DuReader,
respectively.
| new_dataset | 0.715175 |
2305.07507 | Ilias Chalkidis | Ilias Chalkidis, Nicolas Garneau, Catalina Goanta, Daniel Martin Katz,
Anders S{\o}gaard | LeXFiles and LegalLAMA: Facilitating English Multinational Legal
Language Model Development | 9 pages, long paper at ACL 2023 proceedings | null | null | null | cs.CL | http://creativecommons.org/licenses/by-nc-sa/4.0/ | In this work, we conduct a detailed analysis on the performance of
legal-oriented pre-trained language models (PLMs). We examine the interplay
between their original objective, acquired knowledge, and legal language
understanding capacities which we define as the upstream, probing, and
downstream performance, respectively. We consider not only the models' size but
also the pre-training corpora used as important dimensions in our study. To
this end, we release a multinational English legal corpus (LeXFiles) and a
legal knowledge probing benchmark (LegalLAMA) to facilitate training and
detailed analysis of legal-oriented PLMs. We release two new legal PLMs trained
on LeXFiles and evaluate them alongside others on LegalLAMA and LexGLUE. We
find that probing performance strongly correlates with upstream performance in
related legal topics. On the other hand, downstream performance is mainly
driven by the model's size and prior legal knowledge which can be estimated by
upstream and probing performance. Based on these findings, we can conclude that
both dimensions are important for those seeking the development of
domain-specific PLMs.
| [
{
"version": "v1",
"created": "Fri, 12 May 2023 14:21:38 GMT"
},
{
"version": "v2",
"created": "Mon, 22 May 2023 18:20:54 GMT"
}
] | 2023-05-24T00:00:00 | [
[
"Chalkidis",
"Ilias",
""
],
[
"Garneau",
"Nicolas",
""
],
[
"Goanta",
"Catalina",
""
],
[
"Katz",
"Daniel Martin",
""
],
[
"Søgaard",
"Anders",
""
]
] | TITLE: LeXFiles and LegalLAMA: Facilitating English Multinational Legal
Language Model Development
ABSTRACT: In this work, we conduct a detailed analysis on the performance of
legal-oriented pre-trained language models (PLMs). We examine the interplay
between their original objective, acquired knowledge, and legal language
understanding capacities which we define as the upstream, probing, and
downstream performance, respectively. We consider not only the models' size but
also the pre-training corpora used as important dimensions in our study. To
this end, we release a multinational English legal corpus (LeXFiles) and a
legal knowledge probing benchmark (LegalLAMA) to facilitate training and
detailed analysis of legal-oriented PLMs. We release two new legal PLMs trained
on LeXFiles and evaluate them alongside others on LegalLAMA and LexGLUE. We
find that probing performance strongly correlates with upstream performance in
related legal topics. On the other hand, downstream performance is mainly
driven by the model's size and prior legal knowledge which can be estimated by
upstream and probing performance. Based on these findings, we can conclude that
both dimensions are important for those seeking the development of
domain-specific PLMs.
| new_dataset | 0.7116 |
2305.08569 | Jiadong Yu | Jiadong Yu, Ahmad Alhilal, Tailin Zhou, Pan Hui, Danny H.K. Tsang | Attention-based QoE-aware Digital Twin Empowered Edge Computing for
Immersive Virtual Reality | null | null | null | null | eess.SY cs.SY | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Metaverse applications such as virtual reality (VR) content streaming,
require optimal resource allocation strategies for mobile edge computing (MEC)
to ensure a high-quality user experience. In contrast to online reinforcement
learning (RL) algorithms, which can incur substantial communication overheads
and longer delays, the majority of existing works employ offline-trained RL
algorithms for resource allocation decisions in MEC systems. However, they
neglect the impact of desynchronization between the physical and digital worlds
on the effectiveness of the allocation strategy. In this paper, we tackle this
desynchronization using a continual RL framework that facilitates the resource
allocation dynamically for MEC-enabled VR content streaming. We first design a
digital twin-empowered edge computing (DTEC) system and formulate a quality of
experience (QoE) maximization problem based on attention-based resolution
perception. This problem optimizes the allocation of computing and bandwidth
resources while adapting the attention-based resolution of the VR content. The
continual RL framework in DTEC enables adaptive online execution in a
time-varying environment. The reward function is defined based on the QoE and
horizon-fairness QoE (hfQoE) constraints. Furthermore, we propose freshness
prioritized experience replay - continual deep deterministic policy gradient
(FPER-CDDPG) to enhance the performance of continual learning in the presence
of time-varying DT updates. We test FPER-CDDPG using extensive experiments and
evaluation. FPER-CDDPG outperforms the benchmarks in terms of average latency,
QoE, and successful delivery rate as well as meeting the hfQoE requirements and
performance over long-term execution while ensuring system scalability with the
increasing number of users.
| [
{
"version": "v1",
"created": "Mon, 15 May 2023 11:54:46 GMT"
},
{
"version": "v2",
"created": "Tue, 23 May 2023 07:07:33 GMT"
}
] | 2023-05-24T00:00:00 | [
[
"Yu",
"Jiadong",
""
],
[
"Alhilal",
"Ahmad",
""
],
[
"Zhou",
"Tailin",
""
],
[
"Hui",
"Pan",
""
],
[
"Tsang",
"Danny H. K.",
""
]
] | TITLE: Attention-based QoE-aware Digital Twin Empowered Edge Computing for
Immersive Virtual Reality
ABSTRACT: Metaverse applications such as virtual reality (VR) content streaming,
require optimal resource allocation strategies for mobile edge computing (MEC)
to ensure a high-quality user experience. In contrast to online reinforcement
learning (RL) algorithms, which can incur substantial communication overheads
and longer delays, the majority of existing works employ offline-trained RL
algorithms for resource allocation decisions in MEC systems. However, they
neglect the impact of desynchronization between the physical and digital worlds
on the effectiveness of the allocation strategy. In this paper, we tackle this
desynchronization using a continual RL framework that facilitates the resource
allocation dynamically for MEC-enabled VR content streaming. We first design a
digital twin-empowered edge computing (DTEC) system and formulate a quality of
experience (QoE) maximization problem based on attention-based resolution
perception. This problem optimizes the allocation of computing and bandwidth
resources while adapting the attention-based resolution of the VR content. The
continual RL framework in DTEC enables adaptive online execution in a
time-varying environment. The reward function is defined based on the QoE and
horizon-fairness QoE (hfQoE) constraints. Furthermore, we propose freshness
prioritized experience replay - continual deep deterministic policy gradient
(FPER-CDDPG) to enhance the performance of continual learning in the presence
of time-varying DT updates. We test FPER-CDDPG using extensive experiments and
evaluation. FPER-CDDPG outperforms the benchmarks in terms of average latency,
QoE, and successful delivery rate as well as meeting the hfQoE requirements and
performance over long-term execution while ensuring system scalability with the
increasing number of users.
| no_new_dataset | 0.709982 |
2305.10588 | Carina Kauf | Carina Kauf and Anna Ivanova | A Better Way to Do Masked Language Model Scoring | null | null | null | null | cs.CL | http://creativecommons.org/licenses/by/4.0/ | Estimating the log-likelihood of a given sentence under an autoregressive
language model is straightforward: one can simply apply the chain rule and sum
the log-likelihood values for each successive token. However, for masked
language models (MLMs), there is no direct way to estimate the log-likelihood
of a sentence. To address this issue, Salazar et al. (2020) propose to estimate
sentence pseudo-log-likelihood (PLL) scores, computed by successively masking
each sentence token, retrieving its score using the rest of the sentence as
context, and summing the resulting values. Here, we demonstrate that the
original PLL method yields inflated scores for out-of-vocabulary words and
propose an adapted metric, in which we mask not only the target token, but also
all within-word tokens to the right of the target. We show that our adapted
metric (PLL-word-l2r) outperforms both the original PLL metric and a PLL metric
in which all within-word tokens are masked. In particular, it better satisfies
theoretical desiderata and better correlates with scores from autoregressive
models. Finally, we show that the choice of metric affects even tightly
controlled, minimal pair evaluation benchmarks (such as BLiMP), underscoring
the importance of selecting an appropriate scoring metric for evaluating MLM
properties.
| [
{
"version": "v1",
"created": "Wed, 17 May 2023 21:51:58 GMT"
},
{
"version": "v2",
"created": "Tue, 23 May 2023 15:12:33 GMT"
}
] | 2023-05-24T00:00:00 | [
[
"Kauf",
"Carina",
""
],
[
"Ivanova",
"Anna",
""
]
] | TITLE: A Better Way to Do Masked Language Model Scoring
ABSTRACT: Estimating the log-likelihood of a given sentence under an autoregressive
language model is straightforward: one can simply apply the chain rule and sum
the log-likelihood values for each successive token. However, for masked
language models (MLMs), there is no direct way to estimate the log-likelihood
of a sentence. To address this issue, Salazar et al. (2020) propose to estimate
sentence pseudo-log-likelihood (PLL) scores, computed by successively masking
each sentence token, retrieving its score using the rest of the sentence as
context, and summing the resulting values. Here, we demonstrate that the
original PLL method yields inflated scores for out-of-vocabulary words and
propose an adapted metric, in which we mask not only the target token, but also
all within-word tokens to the right of the target. We show that our adapted
metric (PLL-word-l2r) outperforms both the original PLL metric and a PLL metric
in which all within-word tokens are masked. In particular, it better satisfies
theoretical desiderata and better correlates with scores from autoregressive
models. Finally, we show that the choice of metric affects even tightly
controlled, minimal pair evaluation benchmarks (such as BLiMP), underscoring
the importance of selecting an appropriate scoring metric for evaluating MLM
properties.
| no_new_dataset | 0.710459 |
2305.11487 | Guangyan Chen | Guangyan Chen, Meiling Wang, Yi Yang, Kai Yu, Li Yuan, Yufeng Yue | PointGPT: Auto-regressively Generative Pre-training from Point Clouds | 9 pages, 2 figures | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Large language models (LLMs) based on the generative pre-training transformer
(GPT) have demonstrated remarkable effectiveness across a diverse range of
downstream tasks. Inspired by the advancements of the GPT, we present PointGPT,
a novel approach that extends the concept of GPT to point clouds, addressing
the challenges associated with disorder properties, low information density,
and task gaps. Specifically, a point cloud auto-regressive generation task is
proposed to pre-train transformer models. Our method partitions the input point
cloud into multiple point patches and arranges them in an ordered sequence
based on their spatial proximity. Then, an extractor-generator based
transformer decoder, with a dual masking strategy, learns latent
representations conditioned on the preceding point patches, aiming to predict
the next one in an auto-regressive manner. Our scalable approach allows for
learning high-capacity models that generalize well, achieving state-of-the-art
performance on various downstream tasks. In particular, our approach achieves
classification accuracies of 94.9% on the ModelNet40 dataset and 93.4% on the
ScanObjectNN dataset, outperforming all other transformer models. Furthermore,
our method also attains new state-of-the-art accuracies on all four few-shot
learning benchmarks.
| [
{
"version": "v1",
"created": "Fri, 19 May 2023 07:39:04 GMT"
},
{
"version": "v2",
"created": "Tue, 23 May 2023 02:38:26 GMT"
}
] | 2023-05-24T00:00:00 | [
[
"Chen",
"Guangyan",
""
],
[
"Wang",
"Meiling",
""
],
[
"Yang",
"Yi",
""
],
[
"Yu",
"Kai",
""
],
[
"Yuan",
"Li",
""
],
[
"Yue",
"Yufeng",
""
]
] | TITLE: PointGPT: Auto-regressively Generative Pre-training from Point Clouds
ABSTRACT: Large language models (LLMs) based on the generative pre-training transformer
(GPT) have demonstrated remarkable effectiveness across a diverse range of
downstream tasks. Inspired by the advancements of the GPT, we present PointGPT,
a novel approach that extends the concept of GPT to point clouds, addressing
the challenges associated with disorder properties, low information density,
and task gaps. Specifically, a point cloud auto-regressive generation task is
proposed to pre-train transformer models. Our method partitions the input point
cloud into multiple point patches and arranges them in an ordered sequence
based on their spatial proximity. Then, an extractor-generator based
transformer decoder, with a dual masking strategy, learns latent
representations conditioned on the preceding point patches, aiming to predict
the next one in an auto-regressive manner. Our scalable approach allows for
learning high-capacity models that generalize well, achieving state-of-the-art
performance on various downstream tasks. In particular, our approach achieves
classification accuracies of 94.9% on the ModelNet40 dataset and 93.4% on the
ScanObjectNN dataset, outperforming all other transformer models. Furthermore,
our method also attains new state-of-the-art accuracies on all four few-shot
learning benchmarks.
| no_new_dataset | 0.710038 |
2305.12223 | Guangzhi Wang | Guangzhi Wang, Yixiao Ge, Xiaohan Ding, Mohan Kankanhalli, Ying Shan | What Makes for Good Visual Tokenizers for Large Language Models? | 15 pages, 3 figures. Project released at:
https://github.com/TencentARC/GVT | null | null | 01 | cs.CV | http://creativecommons.org/licenses/by/4.0/ | We empirically investigate proper pre-training methods to build good visual
tokenizers, making Large Language Models (LLMs) powerful Multimodal Large
Language Models (MLLMs). In our benchmark, which is curated to evaluate MLLMs
visual semantic understanding and fine-grained perception capabilities, we
discussed different visual tokenizers pre-trained with dominant methods (i.e.,
DeiT, CLIP, MAE, DINO), and observe that: i) Fully/weakly supervised models
capture more semantics than self-supervised models, but the gap is narrowed by
scaling up the pre-training dataset. ii) Self-supervised models are better at
fine-grained perception, where patch-level supervision is particularly
effective. iii) Tuning the visual tokenizer leads to the loss of semantics
obtained from large-scale pretraining, which is unfavorable with relatively
small-scale instruction-tuning dataset. Given the findings, we reviewed methods
that attempted to unify semantics and fine-grained visual understanding, e.g.,
patch-level feature distillation with semantically-rich targets. We obtain an
intriguing insight mask-based strategies that were once all the rage may not be
applicable for obtaining good visual tokenizers. Based on this critical
observation, we obtain a new MLLM equipped with a tailored Good Visual
Tokenizer (GVT), which exhibits strong visual comprehension capability at
multiple scales. In particular, without introducing extra parameters and
task-specific fine-tuning, GVT achieves superior performance on visual question
answering, image captioning, and other fine-grained visual understanding tasks
such as object counting and multi-class identification.
| [
{
"version": "v1",
"created": "Sat, 20 May 2023 16:11:26 GMT"
},
{
"version": "v2",
"created": "Tue, 23 May 2023 10:35:35 GMT"
}
] | 2023-05-24T00:00:00 | [
[
"Wang",
"Guangzhi",
""
],
[
"Ge",
"Yixiao",
""
],
[
"Ding",
"Xiaohan",
""
],
[
"Kankanhalli",
"Mohan",
""
],
[
"Shan",
"Ying",
""
]
] | TITLE: What Makes for Good Visual Tokenizers for Large Language Models?
ABSTRACT: We empirically investigate proper pre-training methods to build good visual
tokenizers, making Large Language Models (LLMs) powerful Multimodal Large
Language Models (MLLMs). In our benchmark, which is curated to evaluate MLLMs
visual semantic understanding and fine-grained perception capabilities, we
discussed different visual tokenizers pre-trained with dominant methods (i.e.,
DeiT, CLIP, MAE, DINO), and observe that: i) Fully/weakly supervised models
capture more semantics than self-supervised models, but the gap is narrowed by
scaling up the pre-training dataset. ii) Self-supervised models are better at
fine-grained perception, where patch-level supervision is particularly
effective. iii) Tuning the visual tokenizer leads to the loss of semantics
obtained from large-scale pretraining, which is unfavorable with relatively
small-scale instruction-tuning dataset. Given the findings, we reviewed methods
that attempted to unify semantics and fine-grained visual understanding, e.g.,
patch-level feature distillation with semantically-rich targets. We obtain an
intriguing insight mask-based strategies that were once all the rage may not be
applicable for obtaining good visual tokenizers. Based on this critical
observation, we obtain a new MLLM equipped with a tailored Good Visual
Tokenizer (GVT), which exhibits strong visual comprehension capability at
multiple scales. In particular, without introducing extra parameters and
task-specific fine-tuning, GVT achieves superior performance on visual question
answering, image captioning, and other fine-grained visual understanding tasks
such as object counting and multi-class identification.
| no_new_dataset | 0.71022 |
2305.12296 | Xilong Zhou | Xilong Zhou, Milo\v{s} Ha\v{s}an, Valentin Deschaintre, Paul Guerrero,
Yannick Hold-Geoffroy, Kalyan Sunkavalli, Nima Khademi Kalantari | PhotoMat: A Material Generator Learned from Single Flash Photos | null | Siggraph 2023 | 10.1145/3588432.3591535 | null | cs.CV cs.AI cs.GR | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Authoring high-quality digital materials is key to realism in 3D rendering.
Previous generative models for materials have been trained exclusively on
synthetic data; such data is limited in availability and has a visual gap to
real materials. We circumvent this limitation by proposing PhotoMat: the first
material generator trained exclusively on real photos of material samples
captured using a cell phone camera with flash. Supervision on individual
material maps is not available in this setting. Instead, we train a generator
for a neural material representation that is rendered with a learned relighting
module to create arbitrarily lit RGB images; these are compared against real
photos using a discriminator. We then train a material maps estimator to decode
material reflectance properties from the neural material representation. We
train PhotoMat with a new dataset of 12,000 material photos captured with
handheld phone cameras under flash lighting. We demonstrate that our generated
materials have better visual quality than previous material generators trained
on synthetic data. Moreover, we can fit analytical material models to closely
match these generated neural materials, thus allowing for further editing and
use in 3D rendering.
| [
{
"version": "v1",
"created": "Sat, 20 May 2023 22:27:41 GMT"
},
{
"version": "v2",
"created": "Tue, 23 May 2023 17:26:27 GMT"
}
] | 2023-05-24T00:00:00 | [
[
"Zhou",
"Xilong",
""
],
[
"Hašan",
"Miloš",
""
],
[
"Deschaintre",
"Valentin",
""
],
[
"Guerrero",
"Paul",
""
],
[
"Hold-Geoffroy",
"Yannick",
""
],
[
"Sunkavalli",
"Kalyan",
""
],
[
"Kalantari",
"Nima Khademi",
""
]
] | TITLE: PhotoMat: A Material Generator Learned from Single Flash Photos
ABSTRACT: Authoring high-quality digital materials is key to realism in 3D rendering.
Previous generative models for materials have been trained exclusively on
synthetic data; such data is limited in availability and has a visual gap to
real materials. We circumvent this limitation by proposing PhotoMat: the first
material generator trained exclusively on real photos of material samples
captured using a cell phone camera with flash. Supervision on individual
material maps is not available in this setting. Instead, we train a generator
for a neural material representation that is rendered with a learned relighting
module to create arbitrarily lit RGB images; these are compared against real
photos using a discriminator. We then train a material maps estimator to decode
material reflectance properties from the neural material representation. We
train PhotoMat with a new dataset of 12,000 material photos captured with
handheld phone cameras under flash lighting. We demonstrate that our generated
materials have better visual quality than previous material generators trained
on synthetic data. Moreover, we can fit analytical material models to closely
match these generated neural materials, thus allowing for further editing and
use in 3D rendering.
| new_dataset | 0.715598 |
2305.12380 | Dario Zanca | Dario Zanca, Andrea Zugarini, Simon Dietz, Thomas R. Altstidl, Mark A.
Turban Ndjeuha, Leo Schwinn, Bjoern Eskofier | Contrastive Language-Image Pretrained Models are Zero-Shot Human
Scanpath Predictors | null | null | null | null | cs.CV cs.AI | http://creativecommons.org/licenses/by/4.0/ | Understanding the mechanisms underlying human attention is a fundamental
challenge for both vision science and artificial intelligence. While numerous
computational models of free-viewing have been proposed, less is known about
the mechanisms underlying task-driven image exploration. To address this gap,
we present CapMIT1003, a database of captions and click-contingent image
explorations collected during captioning tasks. CapMIT1003 is based on the same
stimuli from the well-known MIT1003 benchmark, for which eye-tracking data
under free-viewing conditions is available, which offers a promising
opportunity to concurrently study human attention under both tasks. We make
this dataset publicly available to facilitate future research in this field. In
addition, we introduce NevaClip, a novel zero-shot method for predicting visual
scanpaths that combines contrastive language-image pretrained (CLIP) models
with biologically-inspired neural visual attention (NeVA) algorithms. NevaClip
simulates human scanpaths by aligning the representation of the foveated visual
stimulus and the representation of the associated caption, employing
gradient-driven visual exploration to generate scanpaths. Our experimental
results demonstrate that NevaClip outperforms existing unsupervised
computational models of human visual attention in terms of scanpath
plausibility, for both captioning and free-viewing tasks. Furthermore, we show
that conditioning NevaClip with incorrect or misleading captions leads to
random behavior, highlighting the significant impact of caption guidance in the
decision-making process. These findings contribute to a better understanding of
mechanisms that guide human attention and pave the way for more sophisticated
computational approaches to scanpath prediction that can integrate direct
top-down guidance of downstream tasks.
| [
{
"version": "v1",
"created": "Sun, 21 May 2023 07:24:50 GMT"
},
{
"version": "v2",
"created": "Tue, 23 May 2023 11:17:32 GMT"
}
] | 2023-05-24T00:00:00 | [
[
"Zanca",
"Dario",
""
],
[
"Zugarini",
"Andrea",
""
],
[
"Dietz",
"Simon",
""
],
[
"Altstidl",
"Thomas R.",
""
],
[
"Ndjeuha",
"Mark A. Turban",
""
],
[
"Schwinn",
"Leo",
""
],
[
"Eskofier",
"Bjoern",
""
]
] | TITLE: Contrastive Language-Image Pretrained Models are Zero-Shot Human
Scanpath Predictors
ABSTRACT: Understanding the mechanisms underlying human attention is a fundamental
challenge for both vision science and artificial intelligence. While numerous
computational models of free-viewing have been proposed, less is known about
the mechanisms underlying task-driven image exploration. To address this gap,
we present CapMIT1003, a database of captions and click-contingent image
explorations collected during captioning tasks. CapMIT1003 is based on the same
stimuli from the well-known MIT1003 benchmark, for which eye-tracking data
under free-viewing conditions is available, which offers a promising
opportunity to concurrently study human attention under both tasks. We make
this dataset publicly available to facilitate future research in this field. In
addition, we introduce NevaClip, a novel zero-shot method for predicting visual
scanpaths that combines contrastive language-image pretrained (CLIP) models
with biologically-inspired neural visual attention (NeVA) algorithms. NevaClip
simulates human scanpaths by aligning the representation of the foveated visual
stimulus and the representation of the associated caption, employing
gradient-driven visual exploration to generate scanpaths. Our experimental
results demonstrate that NevaClip outperforms existing unsupervised
computational models of human visual attention in terms of scanpath
plausibility, for both captioning and free-viewing tasks. Furthermore, we show
that conditioning NevaClip with incorrect or misleading captions leads to
random behavior, highlighting the significant impact of caption guidance in the
decision-making process. These findings contribute to a better understanding of
mechanisms that guide human attention and pave the way for more sophisticated
computational approaches to scanpath prediction that can integrate direct
top-down guidance of downstream tasks.
| new_dataset | 0.715809 |
2305.12542 | Zachary Yang | Zachary Yang, Yasmine Maricar, MohammadReza Davari, Nicolas
Grenon-Godbout, Reihaneh Rabbany | ToxBuster: In-game Chat Toxicity Buster with BERT | 11 pages, 3 figures | null | null | null | cs.CL cs.CY | http://creativecommons.org/licenses/by-sa/4.0/ | Detecting toxicity in online spaces is challenging and an ever more pressing
problem given the increase in social media and gaming consumption. We introduce
ToxBuster, a simple and scalable model trained on a relatively large dataset of
194k lines of game chat from Rainbow Six Siege and For Honor, carefully
annotated for different kinds of toxicity. Compared to the existing
state-of-the-art, ToxBuster achieves 82.95% (+7) in precision and 83.56% (+57)
in recall. This improvement is obtained by leveraging past chat history and
metadata. We also study the implication towards real-time and post-game
moderation as well as the model transferability from one game to another.
| [
{
"version": "v1",
"created": "Sun, 21 May 2023 18:53:26 GMT"
}
] | 2023-05-24T00:00:00 | [
[
"Yang",
"Zachary",
""
],
[
"Maricar",
"Yasmine",
""
],
[
"Davari",
"MohammadReza",
""
],
[
"Grenon-Godbout",
"Nicolas",
""
],
[
"Rabbany",
"Reihaneh",
""
]
] | TITLE: ToxBuster: In-game Chat Toxicity Buster with BERT
ABSTRACT: Detecting toxicity in online spaces is challenging and an ever more pressing
problem given the increase in social media and gaming consumption. We introduce
ToxBuster, a simple and scalable model trained on a relatively large dataset of
194k lines of game chat from Rainbow Six Siege and For Honor, carefully
annotated for different kinds of toxicity. Compared to the existing
state-of-the-art, ToxBuster achieves 82.95% (+7) in precision and 83.56% (+57)
in recall. This improvement is obtained by leveraging past chat history and
metadata. We also study the implication towards real-time and post-game
moderation as well as the model transferability from one game to another.
| no_new_dataset | 0.64131 |
2305.13292 | Chen Guo | Guo Chen, Yin-Dong Zheng, Jiahao Wang, Jilan Xu, Yifei Huang, Junting
Pan, Yi Wang, Yali Wang, Yu Qiao, Tong Lu, Limin Wang | VideoLLM: Modeling Video Sequence with Large Language Models | Technical Report | null | null | null | cs.CV | http://creativecommons.org/licenses/by-nc-nd/4.0/ | With the exponential growth of video data, there is an urgent need for
automated technology to analyze and comprehend video content. However, existing
video understanding models are often task-specific and lack a comprehensive
capability of handling diverse tasks. The success of large language models
(LLMs) like GPT has demonstrated their impressive abilities in sequence causal
reasoning. Building upon this insight, we propose a novel framework called
VideoLLM that leverages the sequence reasoning capabilities of pre-trained LLMs
from natural language processing (NLP) for video sequence understanding.
VideoLLM incorporates a carefully designed Modality Encoder and Semantic
Translator, which convert inputs from various modalities into a unified token
sequence. This token sequence is then fed into a decoder-only LLM.
Subsequently, with the aid of a simple task head, our VideoLLM yields an
effective unified framework for different kinds of video understanding tasks.
To evaluate the efficacy of VideoLLM, we conduct extensive experiments using
multiple LLMs and fine-tuning methods. We evaluate our VideoLLM on eight tasks
sourced from four different datasets. The experimental results demonstrate that
the understanding and reasoning capabilities of LLMs can be effectively
transferred to video understanding tasks. We release the code at
https://github.com/cg1177/VideoLLM.
| [
{
"version": "v1",
"created": "Mon, 22 May 2023 17:51:22 GMT"
},
{
"version": "v2",
"created": "Tue, 23 May 2023 07:48:15 GMT"
}
] | 2023-05-24T00:00:00 | [
[
"Chen",
"Guo",
""
],
[
"Zheng",
"Yin-Dong",
""
],
[
"Wang",
"Jiahao",
""
],
[
"Xu",
"Jilan",
""
],
[
"Huang",
"Yifei",
""
],
[
"Pan",
"Junting",
""
],
[
"Wang",
"Yi",
""
],
[
"Wang",
"Yali",
""
],
[
"Qiao",
"Yu",
""
],
[
"Lu",
"Tong",
""
],
[
"Wang",
"Limin",
""
]
] | TITLE: VideoLLM: Modeling Video Sequence with Large Language Models
ABSTRACT: With the exponential growth of video data, there is an urgent need for
automated technology to analyze and comprehend video content. However, existing
video understanding models are often task-specific and lack a comprehensive
capability of handling diverse tasks. The success of large language models
(LLMs) like GPT has demonstrated their impressive abilities in sequence causal
reasoning. Building upon this insight, we propose a novel framework called
VideoLLM that leverages the sequence reasoning capabilities of pre-trained LLMs
from natural language processing (NLP) for video sequence understanding.
VideoLLM incorporates a carefully designed Modality Encoder and Semantic
Translator, which convert inputs from various modalities into a unified token
sequence. This token sequence is then fed into a decoder-only LLM.
Subsequently, with the aid of a simple task head, our VideoLLM yields an
effective unified framework for different kinds of video understanding tasks.
To evaluate the efficacy of VideoLLM, we conduct extensive experiments using
multiple LLMs and fine-tuning methods. We evaluate our VideoLLM on eight tasks
sourced from four different datasets. The experimental results demonstrate that
the understanding and reasoning capabilities of LLMs can be effectively
transferred to video understanding tasks. We release the code at
https://github.com/cg1177/VideoLLM.
| no_new_dataset | 0.708767 |
2305.13315 | Zhao Xu | Zhao Xu, Yaochen Xie, Youzhi Luo, Xuan Zhang, Xinyi Xu, Meng Liu,
Kaleb Dickerson, Cheng Deng, Maho Nakata, Shuiwang Ji | 3D Molecular Geometry Analysis with 2D Graphs | null | null | null | null | physics.chem-ph cs.AI cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Ground-state 3D geometries of molecules are essential for many molecular
analysis tasks. Modern quantum mechanical methods can compute accurate 3D
geometries but are computationally prohibitive. Currently, an efficient
alternative to computing ground-state 3D molecular geometries from 2D graphs is
lacking. Here, we propose a novel deep learning framework to predict 3D
geometries from molecular graphs. To this end, we develop an equilibrium
message passing neural network (EMPNN) to better capture ground-state
geometries from molecular graphs. To provide a testbed for 3D molecular
geometry analysis, we develop a benchmark that includes a large-scale molecular
geometry dataset, data splits, and evaluation protocols. Experimental results
show that EMPNN can efficiently predict more accurate ground-state 3D
geometries than RDKit and other deep learning methods. Results also show that
the proposed framework outperforms self-supervised learning methods on property
prediction tasks.
| [
{
"version": "v1",
"created": "Mon, 1 May 2023 19:00:46 GMT"
}
] | 2023-05-24T00:00:00 | [
[
"Xu",
"Zhao",
""
],
[
"Xie",
"Yaochen",
""
],
[
"Luo",
"Youzhi",
""
],
[
"Zhang",
"Xuan",
""
],
[
"Xu",
"Xinyi",
""
],
[
"Liu",
"Meng",
""
],
[
"Dickerson",
"Kaleb",
""
],
[
"Deng",
"Cheng",
""
],
[
"Nakata",
"Maho",
""
],
[
"Ji",
"Shuiwang",
""
]
] | TITLE: 3D Molecular Geometry Analysis with 2D Graphs
ABSTRACT: Ground-state 3D geometries of molecules are essential for many molecular
analysis tasks. Modern quantum mechanical methods can compute accurate 3D
geometries but are computationally prohibitive. Currently, an efficient
alternative to computing ground-state 3D molecular geometries from 2D graphs is
lacking. Here, we propose a novel deep learning framework to predict 3D
geometries from molecular graphs. To this end, we develop an equilibrium
message passing neural network (EMPNN) to better capture ground-state
geometries from molecular graphs. To provide a testbed for 3D molecular
geometry analysis, we develop a benchmark that includes a large-scale molecular
geometry dataset, data splits, and evaluation protocols. Experimental results
show that EMPNN can efficiently predict more accurate ground-state 3D
geometries than RDKit and other deep learning methods. Results also show that
the proposed framework outperforms self-supervised learning methods on property
prediction tasks.
| new_dataset | 0.717822 |
2305.13317 | Atriya Sen | Rasha Alshawi, Atriya Sen, Nathan S. Upham, Beckett Sterner | A Novel Dataset Towards Extracting Virus-Host Interactions | for associated dataset, see
https://1drv.ms/u/s!AjanaqjkM2-ShN9rMT1BqrpAQRpakQ?e=TPkFgh | null | null | null | cs.CL | http://creativecommons.org/licenses/by/4.0/ | We describe a novel dataset for the automated recognition of named taxonomic
and other entities relevant to the association of viruses with their hosts. We
further describe some initial results using pre-trained models on the
named-entity recognition (NER) task on this novel dataset. We propose that our
dataset of manually annotated abstracts now offers a Gold Standard Corpus for
training future NER models in the automated extraction of host-pathogen
detection methods from scientific publications, and further explain how our
work makes first steps towards predicting the important human health-related
concept of viral spillover risk automatically from the scientific literature.
| [
{
"version": "v1",
"created": "Thu, 11 May 2023 17:20:49 GMT"
}
] | 2023-05-24T00:00:00 | [
[
"Alshawi",
"Rasha",
""
],
[
"Sen",
"Atriya",
""
],
[
"Upham",
"Nathan S.",
""
],
[
"Sterner",
"Beckett",
""
]
] | TITLE: A Novel Dataset Towards Extracting Virus-Host Interactions
ABSTRACT: We describe a novel dataset for the automated recognition of named taxonomic
and other entities relevant to the association of viruses with their hosts. We
further describe some initial results using pre-trained models on the
named-entity recognition (NER) task on this novel dataset. We propose that our
dataset of manually annotated abstracts now offers a Gold Standard Corpus for
training future NER models in the automated extraction of host-pathogen
detection methods from scientific publications, and further explain how our
work makes first steps towards predicting the important human health-related
concept of viral spillover risk automatically from the scientific literature.
| new_dataset | 0.711268 |
2305.13331 | Jiyang Tang | Jiyang Tang, William Chen, Xuankai Chang, Shinji Watanabe, Brian
MacWhinney | A New Benchmark of Aphasia Speech Recognition and Detection Based on
E-Branchformer and Multi-task Learning | Accepted at INTERSPEECH 2023. Code: https://github.com/espnet/espnet | null | null | null | eess.AS cs.CL | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Aphasia is a language disorder that affects the speaking ability of millions
of patients. This paper presents a new benchmark for Aphasia speech recognition
and detection tasks using state-of-the-art speech recognition techniques with
the AphsiaBank dataset. Specifically, we introduce two multi-task learning
methods based on the CTC/Attention architecture to perform both tasks
simultaneously. Our system achieves state-of-the-art speaker-level detection
accuracy (97.3%), and a relative WER reduction of 11% for moderate Aphasia
patients. In addition, we demonstrate the generalizability of our approach by
applying it to another disordered speech database, the DementiaBank Pitt
corpus. We will make our all-in-one recipes and pre-trained model publicly
available to facilitate reproducibility. Our standardized data preprocessing
pipeline and open-source recipes enable researchers to compare results
directly, promoting progress in disordered speech processing.
| [
{
"version": "v1",
"created": "Fri, 19 May 2023 15:10:36 GMT"
}
] | 2023-05-24T00:00:00 | [
[
"Tang",
"Jiyang",
""
],
[
"Chen",
"William",
""
],
[
"Chang",
"Xuankai",
""
],
[
"Watanabe",
"Shinji",
""
],
[
"MacWhinney",
"Brian",
""
]
] | TITLE: A New Benchmark of Aphasia Speech Recognition and Detection Based on
E-Branchformer and Multi-task Learning
ABSTRACT: Aphasia is a language disorder that affects the speaking ability of millions
of patients. This paper presents a new benchmark for Aphasia speech recognition
and detection tasks using state-of-the-art speech recognition techniques with
the AphsiaBank dataset. Specifically, we introduce two multi-task learning
methods based on the CTC/Attention architecture to perform both tasks
simultaneously. Our system achieves state-of-the-art speaker-level detection
accuracy (97.3%), and a relative WER reduction of 11% for moderate Aphasia
patients. In addition, we demonstrate the generalizability of our approach by
applying it to another disordered speech database, the DementiaBank Pitt
corpus. We will make our all-in-one recipes and pre-trained model publicly
available to facilitate reproducibility. Our standardized data preprocessing
pipeline and open-source recipes enable researchers to compare results
directly, promoting progress in disordered speech processing.
| no_new_dataset | 0.649995 |
2305.13332 | Michel Meneses | Michel Meneses, Bruno Iwami | Conditional Online Learning for Keyword Spotting | null | null | null | null | eess.AS cs.LG cs.SD | http://creativecommons.org/licenses/by/4.0/ | Modern approaches for keyword spotting rely on training deep neural networks
on large static datasets with i.i.d. distributions. However, the resulting
models tend to underperform when presented with changing data regimes in
real-life applications. This work investigates a simple but effective online
continual learning method that updates a keyword spotter on-device via SGD as
new data becomes available. Contrary to previous research, this work focuses on
learning the same KWS task, which covers most commercial applications. During
experiments with dynamic audio streams in different scenarios, that method
improves the performance of a pre-trained small-footprint model by 34%.
Moreover, experiments demonstrate that, compared to a naive online learning
implementation, conditional model updates based on its performance in a small
hold-out set drawn from the training distribution mitigate catastrophic
forgetting.
| [
{
"version": "v1",
"created": "Fri, 19 May 2023 15:46:31 GMT"
}
] | 2023-05-24T00:00:00 | [
[
"Meneses",
"Michel",
""
],
[
"Iwami",
"Bruno",
""
]
] | TITLE: Conditional Online Learning for Keyword Spotting
ABSTRACT: Modern approaches for keyword spotting rely on training deep neural networks
on large static datasets with i.i.d. distributions. However, the resulting
models tend to underperform when presented with changing data regimes in
real-life applications. This work investigates a simple but effective online
continual learning method that updates a keyword spotter on-device via SGD as
new data becomes available. Contrary to previous research, this work focuses on
learning the same KWS task, which covers most commercial applications. During
experiments with dynamic audio streams in different scenarios, that method
improves the performance of a pre-trained small-footprint model by 34%.
Moreover, experiments demonstrate that, compared to a naive online learning
implementation, conditional model updates based on its performance in a small
hold-out set drawn from the training distribution mitigate catastrophic
forgetting.
| no_new_dataset | 0.711049 |
2305.13333 | Jafar Abdollahi | Jafar Abdollahi | Evaluating LeNet Algorithms in Classification Lung Cancer from
Iraq-Oncology Teaching Hospital/National Center for Cancer Diseases | arXiv admin note: text overlap with arXiv:2106.11342 by other authors | null | null | null | eess.IV cs.CV cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | The advancement of computer-aided detection systems had a significant impact
on clinical analysis and decision-making on human disease. Lung cancer requires
more attention among the numerous diseases being examined because it affects
both men and women, increasing the mortality rate. LeNet, a deep learning
model, is used in this study to detect lung tumors. The studies were run on a
publicly available dataset made up of CT image data (IQ-OTH/NCCD).
Convolutional neural networks (CNNs) were employed in the experiment for
feature extraction and classification. The proposed system was evaluated on
Iraq-Oncology Teaching Hospital/National Center for Cancer Diseases datasets
the success percentage was calculated as 99.51%, sensitivity (93%) and
specificity (95%), and better results were obtained compared to the existing
methods. Development and validation of algorithms such as ours are important
initial steps in the development of software suites that could be adopted in
routine pathological practices and potentially help reduce the burden on
pathologists.
| [
{
"version": "v1",
"created": "Fri, 19 May 2023 19:23:08 GMT"
}
] | 2023-05-24T00:00:00 | [
[
"Abdollahi",
"Jafar",
""
]
] | TITLE: Evaluating LeNet Algorithms in Classification Lung Cancer from
Iraq-Oncology Teaching Hospital/National Center for Cancer Diseases
ABSTRACT: The advancement of computer-aided detection systems had a significant impact
on clinical analysis and decision-making on human disease. Lung cancer requires
more attention among the numerous diseases being examined because it affects
both men and women, increasing the mortality rate. LeNet, a deep learning
model, is used in this study to detect lung tumors. The studies were run on a
publicly available dataset made up of CT image data (IQ-OTH/NCCD).
Convolutional neural networks (CNNs) were employed in the experiment for
feature extraction and classification. The proposed system was evaluated on
Iraq-Oncology Teaching Hospital/National Center for Cancer Diseases datasets
the success percentage was calculated as 99.51%, sensitivity (93%) and
specificity (95%), and better results were obtained compared to the existing
methods. Development and validation of algorithms such as ours are important
initial steps in the development of software suites that could be adopted in
routine pathological practices and potentially help reduce the burden on
pathologists.
| no_new_dataset | 0.707821 |
2305.13342 | Katerina Margatina | Katerina Margatina and Nikolaos Aletras | On the Limitations of Simulating Active Learning | To appear at Findings of ACL 2023 | null | null | null | cs.LG cs.CL | http://creativecommons.org/licenses/by/4.0/ | Active learning (AL) is a human-and-model-in-the-loop paradigm that
iteratively selects informative unlabeled data for human annotation, aiming to
improve over random sampling. However, performing AL experiments with human
annotations on-the-fly is a laborious and expensive process, thus unrealistic
for academic research. An easy fix to this impediment is to simulate AL, by
treating an already labeled and publicly available dataset as the pool of
unlabeled data. In this position paper, we first survey recent literature and
highlight the challenges across all different steps within the AL loop. We
further unveil neglected caveats in the experimental setup that can
significantly affect the quality of AL research. We continue with an
exploration of how the simulation setting can govern empirical findings,
arguing that it might be one of the answers behind the ever posed question
``why do active learning algorithms sometimes fail to outperform random
sampling?''. We argue that evaluating AL algorithms on available labeled
datasets might provide a lower bound as to their effectiveness in real data. We
believe it is essential to collectively shape the best practices for AL
research, particularly as engineering advancements in LLMs push the research
focus towards data-driven approaches (e.g., data efficiency, alignment,
fairness). In light of this, we have developed guidelines for future work. Our
aim is to draw attention to these limitations within the community, in the hope
of finding ways to address them.
| [
{
"version": "v1",
"created": "Sun, 21 May 2023 22:52:13 GMT"
}
] | 2023-05-24T00:00:00 | [
[
"Margatina",
"Katerina",
""
],
[
"Aletras",
"Nikolaos",
""
]
] | TITLE: On the Limitations of Simulating Active Learning
ABSTRACT: Active learning (AL) is a human-and-model-in-the-loop paradigm that
iteratively selects informative unlabeled data for human annotation, aiming to
improve over random sampling. However, performing AL experiments with human
annotations on-the-fly is a laborious and expensive process, thus unrealistic
for academic research. An easy fix to this impediment is to simulate AL, by
treating an already labeled and publicly available dataset as the pool of
unlabeled data. In this position paper, we first survey recent literature and
highlight the challenges across all different steps within the AL loop. We
further unveil neglected caveats in the experimental setup that can
significantly affect the quality of AL research. We continue with an
exploration of how the simulation setting can govern empirical findings,
arguing that it might be one of the answers behind the ever posed question
``why do active learning algorithms sometimes fail to outperform random
sampling?''. We argue that evaluating AL algorithms on available labeled
datasets might provide a lower bound as to their effectiveness in real data. We
believe it is essential to collectively shape the best practices for AL
research, particularly as engineering advancements in LLMs push the research
focus towards data-driven approaches (e.g., data efficiency, alignment,
fairness). In light of this, we have developed guidelines for future work. Our
aim is to draw attention to these limitations within the community, in the hope
of finding ways to address them.
| no_new_dataset | 0.709655 |
2305.13353 | Dongwei Pan | Dongwei Pan, Long Zhuo, Jingtan Piao, Huiwen Luo, Wei Cheng, Yuxin
Wang, Siming Fan, Shengqi Liu, Lei Yang, Bo Dai, Ziwei Liu, Chen Change Loy,
Chen Qian, Wayne Wu, Dahua Lin, Kwan-Yee Lin | RenderMe-360: A Large Digital Asset Library and Benchmarks Towards
High-fidelity Head Avatars | Technical Report; Project Page: 36; Github Link:
https://github.com/RenderMe-360/RenderMe-360 | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Synthesizing high-fidelity head avatars is a central problem for computer
vision and graphics. While head avatar synthesis algorithms have advanced
rapidly, the best ones still face great obstacles in real-world scenarios. One
of the vital causes is inadequate datasets -- 1) current public datasets can
only support researchers to explore high-fidelity head avatars in one or two
task directions; 2) these datasets usually contain digital head assets with
limited data volume, and narrow distribution over different attributes. In this
paper, we present RenderMe-360, a comprehensive 4D human head dataset to drive
advance in head avatar research. It contains massive data assets, with 243+
million complete head frames, and over 800k video sequences from 500 different
identities captured by synchronized multi-view cameras at 30 FPS. It is a
large-scale digital library for head avatars with three key attributes: 1) High
Fidelity: all subjects are captured by 60 synchronized, high-resolution 2K
cameras in 360 degrees. 2) High Diversity: The collected subjects vary from
different ages, eras, ethnicities, and cultures, providing abundant materials
with distinctive styles in appearance and geometry. Moreover, each subject is
asked to perform various motions, such as expressions and head rotations, which
further extend the richness of assets. 3) Rich Annotations: we provide
annotations with different granularities: cameras' parameters, matting, scan,
2D/3D facial landmarks, FLAME fitting, and text description.
Based on the dataset, we build a comprehensive benchmark for head avatar
research, with 16 state-of-the-art methods performed on five main tasks: novel
view synthesis, novel expression synthesis, hair rendering, hair editing, and
talking head generation. Our experiments uncover the strengths and weaknesses
of current methods. RenderMe-360 opens the door for future exploration in head
avatars.
| [
{
"version": "v1",
"created": "Mon, 22 May 2023 17:54:01 GMT"
}
] | 2023-05-24T00:00:00 | [
[
"Pan",
"Dongwei",
""
],
[
"Zhuo",
"Long",
""
],
[
"Piao",
"Jingtan",
""
],
[
"Luo",
"Huiwen",
""
],
[
"Cheng",
"Wei",
""
],
[
"Wang",
"Yuxin",
""
],
[
"Fan",
"Siming",
""
],
[
"Liu",
"Shengqi",
""
],
[
"Yang",
"Lei",
""
],
[
"Dai",
"Bo",
""
],
[
"Liu",
"Ziwei",
""
],
[
"Loy",
"Chen Change",
""
],
[
"Qian",
"Chen",
""
],
[
"Wu",
"Wayne",
""
],
[
"Lin",
"Dahua",
""
],
[
"Lin",
"Kwan-Yee",
""
]
] | TITLE: RenderMe-360: A Large Digital Asset Library and Benchmarks Towards
High-fidelity Head Avatars
ABSTRACT: Synthesizing high-fidelity head avatars is a central problem for computer
vision and graphics. While head avatar synthesis algorithms have advanced
rapidly, the best ones still face great obstacles in real-world scenarios. One
of the vital causes is inadequate datasets -- 1) current public datasets can
only support researchers to explore high-fidelity head avatars in one or two
task directions; 2) these datasets usually contain digital head assets with
limited data volume, and narrow distribution over different attributes. In this
paper, we present RenderMe-360, a comprehensive 4D human head dataset to drive
advance in head avatar research. It contains massive data assets, with 243+
million complete head frames, and over 800k video sequences from 500 different
identities captured by synchronized multi-view cameras at 30 FPS. It is a
large-scale digital library for head avatars with three key attributes: 1) High
Fidelity: all subjects are captured by 60 synchronized, high-resolution 2K
cameras in 360 degrees. 2) High Diversity: The collected subjects vary from
different ages, eras, ethnicities, and cultures, providing abundant materials
with distinctive styles in appearance and geometry. Moreover, each subject is
asked to perform various motions, such as expressions and head rotations, which
further extend the richness of assets. 3) Rich Annotations: we provide
annotations with different granularities: cameras' parameters, matting, scan,
2D/3D facial landmarks, FLAME fitting, and text description.
Based on the dataset, we build a comprehensive benchmark for head avatar
research, with 16 state-of-the-art methods performed on five main tasks: novel
view synthesis, novel expression synthesis, hair rendering, hair editing, and
talking head generation. Our experiments uncover the strengths and weaknesses
of current methods. RenderMe-360 opens the door for future exploration in head
avatars.
| new_dataset | 0.618089 |
2305.13398 | Maysam Orouskhani | Maysam Orouskhani, Negar Firoozeh, Shaojun Xia, Mahmud Mossa-Basha,
Chengcheng Zhu | nnDetection for Intracranial Aneurysms Detection and Localization | 6 pages, 4 figures | null | null | null | cs.CV cs.LG q-bio.QM | http://creativecommons.org/licenses/by/4.0/ | Intracranial aneurysms are a commonly occurring and life-threatening
condition, affecting approximately 3.2% of the general population.
Consequently, detecting these aneurysms plays a crucial role in their
management. Lesion detection involves the simultaneous localization and
categorization of abnormalities within medical images. In this study, we
employed the nnDetection framework, a self-configuring framework specifically
designed for 3D medical object detection, to detect and localize the 3D
coordinates of aneurysms effectively. To capture and extract diverse features
associated with aneurysms, we utilized TOF-MRA and structural MRI, both
obtained from the ADAM dataset. The performance of our proposed deep learning
model was assessed through the utilization of free-response receiver operative
characteristics for evaluation purposes. The model's weights and 3D prediction
of the bounding box of TOF-MRA are publicly available at
https://github.com/orouskhani/AneurysmDetection.
| [
{
"version": "v1",
"created": "Mon, 22 May 2023 18:18:26 GMT"
}
] | 2023-05-24T00:00:00 | [
[
"Orouskhani",
"Maysam",
""
],
[
"Firoozeh",
"Negar",
""
],
[
"Xia",
"Shaojun",
""
],
[
"Mossa-Basha",
"Mahmud",
""
],
[
"Zhu",
"Chengcheng",
""
]
] | TITLE: nnDetection for Intracranial Aneurysms Detection and Localization
ABSTRACT: Intracranial aneurysms are a commonly occurring and life-threatening
condition, affecting approximately 3.2% of the general population.
Consequently, detecting these aneurysms plays a crucial role in their
management. Lesion detection involves the simultaneous localization and
categorization of abnormalities within medical images. In this study, we
employed the nnDetection framework, a self-configuring framework specifically
designed for 3D medical object detection, to detect and localize the 3D
coordinates of aneurysms effectively. To capture and extract diverse features
associated with aneurysms, we utilized TOF-MRA and structural MRI, both
obtained from the ADAM dataset. The performance of our proposed deep learning
model was assessed through the utilization of free-response receiver operative
characteristics for evaluation purposes. The model's weights and 3D prediction
of the bounding box of TOF-MRA are publicly available at
https://github.com/orouskhani/AneurysmDetection.
| no_new_dataset | 0.709849 |
2305.13412 | Yiming Wang | Yiming Wang, Zhuosheng Zhang, Rui Wang | Element-aware Summarization with Large Language Models: Expert-aligned
Evaluation and Chain-of-Thought Method | Accepted by ACL 2023 | null | null | null | cs.CL | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Automatic summarization generates concise summaries that contain key ideas of
source documents. As the most mainstream datasets for the news sub-domain,
CNN/DailyMail and BBC XSum have been widely used for performance benchmarking.
However, the reference summaries of those datasets turn out to be noisy, mainly
in terms of factual hallucination and information redundancy. To address this
challenge, we first annotate new expert-writing Element-aware test sets
following the "Lasswell Communication Model" proposed by Lasswell (1948),
allowing reference summaries to focus on more fine-grained news elements
objectively and comprehensively. Utilizing the new test sets, we observe the
surprising zero-shot summary ability of LLMs, which addresses the issue of the
inconsistent results between human preference and automatic evaluation metrics
of LLMs' zero-shot summaries in prior work. Further, we propose a Summary
Chain-of-Thought (SumCoT) technique to elicit LLMs to generate summaries step
by step, which helps them integrate more fine-grained details of source
documents into the final summaries that correlate with the human writing
mindset. Experimental results show our method outperforms state-of-the-art
fine-tuned PLMs and zero-shot LLMs by +4.33/+4.77 in ROUGE-L on the two
datasets, respectively. Dataset and code are publicly available at
https://github.com/Alsace08/SumCoT.
| [
{
"version": "v1",
"created": "Mon, 22 May 2023 18:54:35 GMT"
}
] | 2023-05-24T00:00:00 | [
[
"Wang",
"Yiming",
""
],
[
"Zhang",
"Zhuosheng",
""
],
[
"Wang",
"Rui",
""
]
] | TITLE: Element-aware Summarization with Large Language Models: Expert-aligned
Evaluation and Chain-of-Thought Method
ABSTRACT: Automatic summarization generates concise summaries that contain key ideas of
source documents. As the most mainstream datasets for the news sub-domain,
CNN/DailyMail and BBC XSum have been widely used for performance benchmarking.
However, the reference summaries of those datasets turn out to be noisy, mainly
in terms of factual hallucination and information redundancy. To address this
challenge, we first annotate new expert-writing Element-aware test sets
following the "Lasswell Communication Model" proposed by Lasswell (1948),
allowing reference summaries to focus on more fine-grained news elements
objectively and comprehensively. Utilizing the new test sets, we observe the
surprising zero-shot summary ability of LLMs, which addresses the issue of the
inconsistent results between human preference and automatic evaluation metrics
of LLMs' zero-shot summaries in prior work. Further, we propose a Summary
Chain-of-Thought (SumCoT) technique to elicit LLMs to generate summaries step
by step, which helps them integrate more fine-grained details of source
documents into the final summaries that correlate with the human writing
mindset. Experimental results show our method outperforms state-of-the-art
fine-tuned PLMs and zero-shot LLMs by +4.33/+4.77 in ROUGE-L on the two
datasets, respectively. Dataset and code are publicly available at
https://github.com/Alsace08/SumCoT.
| no_new_dataset | 0.711243 |
2305.13456 | Isaac Corley | Isaac Corley, Caleb Robinson, Rahul Dodhia, Juan M. Lavista Ferres,
Peyman Najafirad | Revisiting pre-trained remote sensing model benchmarks: resizing and
normalization matters | null | null | null | null | cs.CV cs.LG | http://creativecommons.org/licenses/by/4.0/ | Research in self-supervised learning (SSL) with natural images has progressed
rapidly in recent years and is now increasingly being applied to and
benchmarked with datasets containing remotely sensed imagery. A common
benchmark case is to evaluate SSL pre-trained model embeddings on datasets of
remotely sensed imagery with small patch sizes, e.g., 32x32 pixels, whereas
standard SSL pre-training takes place with larger patch sizes, e.g., 224x224.
Furthermore, pre-training methods tend to use different image normalization
preprocessing steps depending on the dataset. In this paper, we show, across
seven satellite and aerial imagery datasets of varying resolution, that by
simply following the preprocessing steps used in pre-training (precisely, image
sizing and normalization methods), one can achieve significant performance
improvements when evaluating the extracted features on downstream tasks -- an
important detail overlooked in previous work in this space. We show that by
following these steps, ImageNet pre-training remains a competitive baseline for
satellite imagery based transfer learning tasks -- for example we find that
these steps give +32.28 to overall accuracy on the So2Sat random split dataset
and +11.16 on the EuroSAT dataset. Finally, we report comprehensive benchmark
results with a variety of simple baseline methods for each of the seven
datasets, forming an initial benchmark suite for remote sensing imagery.
| [
{
"version": "v1",
"created": "Mon, 22 May 2023 19:57:13 GMT"
}
] | 2023-05-24T00:00:00 | [
[
"Corley",
"Isaac",
""
],
[
"Robinson",
"Caleb",
""
],
[
"Dodhia",
"Rahul",
""
],
[
"Ferres",
"Juan M. Lavista",
""
],
[
"Najafirad",
"Peyman",
""
]
] | TITLE: Revisiting pre-trained remote sensing model benchmarks: resizing and
normalization matters
ABSTRACT: Research in self-supervised learning (SSL) with natural images has progressed
rapidly in recent years and is now increasingly being applied to and
benchmarked with datasets containing remotely sensed imagery. A common
benchmark case is to evaluate SSL pre-trained model embeddings on datasets of
remotely sensed imagery with small patch sizes, e.g., 32x32 pixels, whereas
standard SSL pre-training takes place with larger patch sizes, e.g., 224x224.
Furthermore, pre-training methods tend to use different image normalization
preprocessing steps depending on the dataset. In this paper, we show, across
seven satellite and aerial imagery datasets of varying resolution, that by
simply following the preprocessing steps used in pre-training (precisely, image
sizing and normalization methods), one can achieve significant performance
improvements when evaluating the extracted features on downstream tasks -- an
important detail overlooked in previous work in this space. We show that by
following these steps, ImageNet pre-training remains a competitive baseline for
satellite imagery based transfer learning tasks -- for example we find that
these steps give +32.28 to overall accuracy on the So2Sat random split dataset
and +11.16 on the EuroSAT dataset. Finally, we report comprehensive benchmark
results with a variety of simple baseline methods for each of the seven
datasets, forming an initial benchmark suite for remote sensing imagery.
| no_new_dataset | 0.710622 |
2305.13490 | Vardhan Jai | Jai Vardhan, Kothapalli Sai Swetha | Detection of healthy and diseased crops in drone captured images using
Deep Learning | null | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Monitoring plant health is crucial for maintaining agricultural productivity
and food safety. Disruptions in the plant's normal state, caused by diseases,
often interfere with essential plant activities, and timely detection of these
diseases can significantly mitigate crop loss. In this study, we propose a deep
learning-based approach for efficient detection of plant diseases using
drone-captured imagery. A comprehensive database of various plant species,
exhibiting numerous diseases, was compiled from the Internet and utilized as
the training and test dataset. A Convolutional Neural Network (CNN), renowned
for its performance in image classification tasks, was employed as our primary
predictive model. The CNN model, trained on this rich dataset, demonstrated
superior proficiency in crop disease categorization and detection, even under
challenging imaging conditions. For field implementation, we deployed a
prototype drone model equipped with a high-resolution camera for live
monitoring of extensive agricultural fields. The captured images served as the
input for our trained model, enabling real-time identification of healthy and
diseased plants. Our approach promises an efficient and scalable solution for
improving crop health monitoring systems.
| [
{
"version": "v1",
"created": "Mon, 22 May 2023 21:15:12 GMT"
}
] | 2023-05-24T00:00:00 | [
[
"Vardhan",
"Jai",
""
],
[
"Swetha",
"Kothapalli Sai",
""
]
] | TITLE: Detection of healthy and diseased crops in drone captured images using
Deep Learning
ABSTRACT: Monitoring plant health is crucial for maintaining agricultural productivity
and food safety. Disruptions in the plant's normal state, caused by diseases,
often interfere with essential plant activities, and timely detection of these
diseases can significantly mitigate crop loss. In this study, we propose a deep
learning-based approach for efficient detection of plant diseases using
drone-captured imagery. A comprehensive database of various plant species,
exhibiting numerous diseases, was compiled from the Internet and utilized as
the training and test dataset. A Convolutional Neural Network (CNN), renowned
for its performance in image classification tasks, was employed as our primary
predictive model. The CNN model, trained on this rich dataset, demonstrated
superior proficiency in crop disease categorization and detection, even under
challenging imaging conditions. For field implementation, we deployed a
prototype drone model equipped with a high-resolution camera for live
monitoring of extensive agricultural fields. The captured images served as the
input for our trained model, enabling real-time identification of healthy and
diseased plants. Our approach promises an efficient and scalable solution for
improving crop health monitoring systems.
| no_new_dataset | 0.650745 |
2305.13500 | Sitao Zhang | Sitao Zhang, Yimu Pan, James Z. Wang | Learning Emotion Representations from Verbal and Nonverbal Communication | CVPR 2023 | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Emotion understanding is an essential but highly challenging component of
artificial general intelligence. The absence of extensively annotated datasets
has significantly impeded advancements in this field. We present EmotionCLIP,
the first pre-training paradigm to extract visual emotion representations from
verbal and nonverbal communication using only uncurated data. Compared to
numerical labels or descriptions used in previous methods, communication
naturally contains emotion information. Furthermore, acquiring emotion
representations from communication is more congruent with the human learning
process. We guide EmotionCLIP to attend to nonverbal emotion cues through
subject-aware context encoding and verbal emotion cues using sentiment-guided
contrastive learning. Extensive experiments validate the effectiveness and
transferability of EmotionCLIP. Using merely linear-probe evaluation protocol,
EmotionCLIP outperforms the state-of-the-art supervised visual emotion
recognition methods and rivals many multimodal approaches across various
benchmarks. We anticipate that the advent of EmotionCLIP will address the
prevailing issue of data scarcity in emotion understanding, thereby fostering
progress in related domains. The code and pre-trained models are available at
https://github.com/Xeaver/EmotionCLIP.
| [
{
"version": "v1",
"created": "Mon, 22 May 2023 21:36:55 GMT"
}
] | 2023-05-24T00:00:00 | [
[
"Zhang",
"Sitao",
""
],
[
"Pan",
"Yimu",
""
],
[
"Wang",
"James Z.",
""
]
] | TITLE: Learning Emotion Representations from Verbal and Nonverbal Communication
ABSTRACT: Emotion understanding is an essential but highly challenging component of
artificial general intelligence. The absence of extensively annotated datasets
has significantly impeded advancements in this field. We present EmotionCLIP,
the first pre-training paradigm to extract visual emotion representations from
verbal and nonverbal communication using only uncurated data. Compared to
numerical labels or descriptions used in previous methods, communication
naturally contains emotion information. Furthermore, acquiring emotion
representations from communication is more congruent with the human learning
process. We guide EmotionCLIP to attend to nonverbal emotion cues through
subject-aware context encoding and verbal emotion cues using sentiment-guided
contrastive learning. Extensive experiments validate the effectiveness and
transferability of EmotionCLIP. Using merely linear-probe evaluation protocol,
EmotionCLIP outperforms the state-of-the-art supervised visual emotion
recognition methods and rivals many multimodal approaches across various
benchmarks. We anticipate that the advent of EmotionCLIP will address the
prevailing issue of data scarcity in emotion understanding, thereby fostering
progress in related domains. The code and pre-trained models are available at
https://github.com/Xeaver/EmotionCLIP.
| no_new_dataset | 0.709384 |
2305.13509 | Cuong Ly | Cuong Ly, Grayson Jorgenson, Dan Rosa de Jesus, Henry Kvinge, Adam
Attarian, Yijing Watkins | ColMix -- A Simple Data Augmentation Framework to Improve Object
Detector Performance and Robustness in Aerial Images | null | null | null | null | cs.CV cs.AI cs.LG eess.IV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | In the last decade, Convolutional Neural Network (CNN) and transformer based
object detectors have achieved high performance on a large variety of datasets.
Though the majority of detection literature has developed this capability on
datasets such as MS COCO, these detectors have still proven effective for
remote sensing applications. Challenges in this particular domain, such as
small numbers of annotated objects and low object density, hinder overall
performance. In this work, we present a novel augmentation method, called
collage pasting, for increasing the object density without a need for
segmentation masks, thereby improving the detector performance. We demonstrate
that collage pasting improves precision and recall beyond related methods, such
as mosaic augmentation, and enables greater control of object density. However,
we find that collage pasting is vulnerable to certain out-of-distribution
shifts, such as image corruptions. To address this, we introduce two simple
approaches for combining collage pasting with PixMix augmentation method, and
refer to our combined techniques as ColMix. Through extensive experiments, we
show that employing ColMix results in detectors with superior performance on
aerial imagery datasets and robust to various corruptions.
| [
{
"version": "v1",
"created": "Mon, 22 May 2023 21:56:35 GMT"
}
] | 2023-05-24T00:00:00 | [
[
"Ly",
"Cuong",
""
],
[
"Jorgenson",
"Grayson",
""
],
[
"de Jesus",
"Dan Rosa",
""
],
[
"Kvinge",
"Henry",
""
],
[
"Attarian",
"Adam",
""
],
[
"Watkins",
"Yijing",
""
]
] | TITLE: ColMix -- A Simple Data Augmentation Framework to Improve Object
Detector Performance and Robustness in Aerial Images
ABSTRACT: In the last decade, Convolutional Neural Network (CNN) and transformer based
object detectors have achieved high performance on a large variety of datasets.
Though the majority of detection literature has developed this capability on
datasets such as MS COCO, these detectors have still proven effective for
remote sensing applications. Challenges in this particular domain, such as
small numbers of annotated objects and low object density, hinder overall
performance. In this work, we present a novel augmentation method, called
collage pasting, for increasing the object density without a need for
segmentation masks, thereby improving the detector performance. We demonstrate
that collage pasting improves precision and recall beyond related methods, such
as mosaic augmentation, and enables greater control of object density. However,
we find that collage pasting is vulnerable to certain out-of-distribution
shifts, such as image corruptions. To address this, we introduce two simple
approaches for combining collage pasting with PixMix augmentation method, and
refer to our combined techniques as ColMix. Through extensive experiments, we
show that employing ColMix results in detectors with superior performance on
aerial imagery datasets and robust to various corruptions.
| no_new_dataset | 0.709667 |
2305.13516 | Michael Auli | Vineel Pratap, Andros Tjandra, Bowen Shi, Paden Tomasello, Arun Babu,
Sayani Kundu, Ali Elkahky, Zhaoheng Ni, Apoorv Vyas, Maryam Fazel-Zarandi,
Alexei Baevski, Yossi Adi, Xiaohui Zhang, Wei-Ning Hsu, Alexis Conneau,
Michael Auli | Scaling Speech Technology to 1,000+ Languages | null | null | null | null | cs.CL cs.SD eess.AS | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Expanding the language coverage of speech technology has the potential to
improve access to information for many more people. However, current speech
technology is restricted to about one hundred languages which is a small
fraction of the over 7,000 languages spoken around the world. The Massively
Multilingual Speech (MMS) project increases the number of supported languages
by 10-40x, depending on the task. The main ingredients are a new dataset based
on readings of publicly available religious texts and effectively leveraging
self-supervised learning. We built pre-trained wav2vec 2.0 models covering
1,406 languages, a single multilingual automatic speech recognition model for
1,107 languages, speech synthesis models for the same number of languages, as
well as a language identification model for 4,017 languages. Experiments show
that our multilingual speech recognition model more than halves the word error
rate of Whisper on 54 languages of the FLEURS benchmark while being trained on
a small fraction of the labeled data.
| [
{
"version": "v1",
"created": "Mon, 22 May 2023 22:09:41 GMT"
}
] | 2023-05-24T00:00:00 | [
[
"Pratap",
"Vineel",
""
],
[
"Tjandra",
"Andros",
""
],
[
"Shi",
"Bowen",
""
],
[
"Tomasello",
"Paden",
""
],
[
"Babu",
"Arun",
""
],
[
"Kundu",
"Sayani",
""
],
[
"Elkahky",
"Ali",
""
],
[
"Ni",
"Zhaoheng",
""
],
[
"Vyas",
"Apoorv",
""
],
[
"Fazel-Zarandi",
"Maryam",
""
],
[
"Baevski",
"Alexei",
""
],
[
"Adi",
"Yossi",
""
],
[
"Zhang",
"Xiaohui",
""
],
[
"Hsu",
"Wei-Ning",
""
],
[
"Conneau",
"Alexis",
""
],
[
"Auli",
"Michael",
""
]
] | TITLE: Scaling Speech Technology to 1,000+ Languages
ABSTRACT: Expanding the language coverage of speech technology has the potential to
improve access to information for many more people. However, current speech
technology is restricted to about one hundred languages which is a small
fraction of the over 7,000 languages spoken around the world. The Massively
Multilingual Speech (MMS) project increases the number of supported languages
by 10-40x, depending on the task. The main ingredients are a new dataset based
on readings of publicly available religious texts and effectively leveraging
self-supervised learning. We built pre-trained wav2vec 2.0 models covering
1,406 languages, a single multilingual automatic speech recognition model for
1,107 languages, speech synthesis models for the same number of languages, as
well as a language identification model for 4,017 languages. Experiments show
that our multilingual speech recognition model more than halves the word error
rate of Whisper on 54 languages of the FLEURS benchmark while being trained on
a small fraction of the labeled data.
| new_dataset | 0.717427 |
2305.13534 | Muru Zhang | Muru Zhang, Ofir Press, William Merrill, Alisa Liu, Noah A. Smith | How Language Model Hallucinations Can Snowball | null | null | null | null | cs.CL | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | A major risk of using language models in practical applications is their
tendency to hallucinate incorrect statements. Hallucinations are often
attributed to knowledge gaps in LMs, but we hypothesize that in some cases,
when justifying previously generated hallucinations, LMs output false claims
that they can separately recognize as incorrect. We construct three
question-answering datasets where ChatGPT and GPT-4 often state an incorrect
answer and offer an explanation with at least one incorrect claim. Crucially,
we find that ChatGPT and GPT-4 can identify 67% and 87% of their own mistakes,
respectively. We refer to this phenomenon as hallucination snowballing: an LM
over-commits to early mistakes, leading to more mistakes that it otherwise
would not make.
| [
{
"version": "v1",
"created": "Mon, 22 May 2023 23:14:28 GMT"
}
] | 2023-05-24T00:00:00 | [
[
"Zhang",
"Muru",
""
],
[
"Press",
"Ofir",
""
],
[
"Merrill",
"William",
""
],
[
"Liu",
"Alisa",
""
],
[
"Smith",
"Noah A.",
""
]
] | TITLE: How Language Model Hallucinations Can Snowball
ABSTRACT: A major risk of using language models in practical applications is their
tendency to hallucinate incorrect statements. Hallucinations are often
attributed to knowledge gaps in LMs, but we hypothesize that in some cases,
when justifying previously generated hallucinations, LMs output false claims
that they can separately recognize as incorrect. We construct three
question-answering datasets where ChatGPT and GPT-4 often state an incorrect
answer and offer an explanation with at least one incorrect claim. Crucially,
we find that ChatGPT and GPT-4 can identify 67% and 87% of their own mistakes,
respectively. We refer to this phenomenon as hallucination snowballing: an LM
over-commits to early mistakes, leading to more mistakes that it otherwise
would not make.
| new_dataset | 0.711794 |
2305.13535 | Ananth Balashankar | Ananth Balashankar, Xuezhi Wang, Yao Qin, Ben Packer, Nithum Thain,
Jilin Chen, Ed H. Chi, Alex Beutel | Improving Classifier Robustness through Active Generation of Pairwise
Counterfactuals | null | null | null | null | cs.CL cs.LG | http://creativecommons.org/licenses/by/4.0/ | Counterfactual Data Augmentation (CDA) is a commonly used technique for
improving robustness in natural language classifiers. However, one fundamental
challenge is how to discover meaningful counterfactuals and efficiently label
them, with minimal human labeling cost. Most existing methods either completely
rely on human-annotated labels, an expensive process which limits the scale of
counterfactual data, or implicitly assume label invariance, which may mislead
the model with incorrect labels. In this paper, we present a novel framework
that utilizes counterfactual generative models to generate a large number of
diverse counterfactuals by actively sampling from regions of uncertainty, and
then automatically label them with a learned pairwise classifier. Our key
insight is that we can more correctly label the generated counterfactuals by
training a pairwise classifier that interpolates the relationship between the
original example and the counterfactual. We demonstrate that with a small
amount of human-annotated counterfactual data (10%), we can generate a
counterfactual augmentation dataset with learned labels, that provides an
18-20% improvement in robustness and a 14-21% reduction in errors on 6
out-of-domain datasets, comparable to that of a fully human-annotated
counterfactual dataset for both sentiment classification and question
paraphrase tasks.
| [
{
"version": "v1",
"created": "Mon, 22 May 2023 23:19:01 GMT"
}
] | 2023-05-24T00:00:00 | [
[
"Balashankar",
"Ananth",
""
],
[
"Wang",
"Xuezhi",
""
],
[
"Qin",
"Yao",
""
],
[
"Packer",
"Ben",
""
],
[
"Thain",
"Nithum",
""
],
[
"Chen",
"Jilin",
""
],
[
"Chi",
"Ed H.",
""
],
[
"Beutel",
"Alex",
""
]
] | TITLE: Improving Classifier Robustness through Active Generation of Pairwise
Counterfactuals
ABSTRACT: Counterfactual Data Augmentation (CDA) is a commonly used technique for
improving robustness in natural language classifiers. However, one fundamental
challenge is how to discover meaningful counterfactuals and efficiently label
them, with minimal human labeling cost. Most existing methods either completely
rely on human-annotated labels, an expensive process which limits the scale of
counterfactual data, or implicitly assume label invariance, which may mislead
the model with incorrect labels. In this paper, we present a novel framework
that utilizes counterfactual generative models to generate a large number of
diverse counterfactuals by actively sampling from regions of uncertainty, and
then automatically label them with a learned pairwise classifier. Our key
insight is that we can more correctly label the generated counterfactuals by
training a pairwise classifier that interpolates the relationship between the
original example and the counterfactual. We demonstrate that with a small
amount of human-annotated counterfactual data (10%), we can generate a
counterfactual augmentation dataset with learned labels, that provides an
18-20% improvement in robustness and a 14-21% reduction in errors on 6
out-of-domain datasets, comparable to that of a fully human-annotated
counterfactual dataset for both sentiment classification and question
paraphrase tasks.
| new_dataset | 0.674881 |
2305.13541 | Yu Guan | Shuai Shao, Yu Guan, Bing Zhai, Paolo Missier, Thomas Ploetz | ConvBoost: Boosting ConvNets for Sensor-based Activity Recognition | 21 pages | Proc. ACM Interact. Mob. Wearable Ubiquitous Technol. 7, 2,
Article 75 (June 2023) | 10.1145/3596234 | null | cs.LG cs.AI cs.CV cs.HC | http://creativecommons.org/licenses/by/4.0/ | Human activity recognition (HAR) is one of the core research themes in
ubiquitous and wearable computing. With the shift to deep learning (DL) based
analysis approaches, it has become possible to extract high-level features and
perform classification in an end-to-end manner. Despite their promising overall
capabilities, DL-based HAR may suffer from overfitting due to the notoriously
small, often inadequate, amounts of labeled sample data that are available for
typical HAR applications. In response to such challenges, we propose ConvBoost
-- a novel, three-layer, structured model architecture and boosting framework
for convolutional network based HAR. Our framework generates additional
training data from three different perspectives for improved HAR, aiming to
alleviate the shortness of labeled training data in the field. Specifically,
with the introduction of three conceptual layers--Sampling Layer, Data
Augmentation Layer, and Resilient Layer -- we develop three "boosters" --
R-Frame, Mix-up, and C-Drop -- to enrich the per-epoch training data by
dense-sampling, synthesizing, and simulating, respectively. These new
conceptual layers and boosters, that are universally applicable for any kind of
convolutional network, have been designed based on the characteristics of the
sensor data and the concept of frame-wise HAR. In our experimental evaluation
on three standard benchmarks (Opportunity, PAMAP2, GOTOV) we demonstrate the
effectiveness of our ConvBoost framework for HAR applications based on variants
of convolutional networks: vanilla CNN, ConvLSTM, and Attention Models. We
achieved substantial performance gains for all of them, which suggests that the
proposed approach is generic and can serve as a practical solution for boosting
the performance of existing ConvNet-based HAR models. This is an open-source
project, and the code can be found at https://github.com/sshao2013/ConvBoost
| [
{
"version": "v1",
"created": "Mon, 22 May 2023 23:27:24 GMT"
}
] | 2023-05-24T00:00:00 | [
[
"Shao",
"Shuai",
""
],
[
"Guan",
"Yu",
""
],
[
"Zhai",
"Bing",
""
],
[
"Missier",
"Paolo",
""
],
[
"Ploetz",
"Thomas",
""
]
] | TITLE: ConvBoost: Boosting ConvNets for Sensor-based Activity Recognition
ABSTRACT: Human activity recognition (HAR) is one of the core research themes in
ubiquitous and wearable computing. With the shift to deep learning (DL) based
analysis approaches, it has become possible to extract high-level features and
perform classification in an end-to-end manner. Despite their promising overall
capabilities, DL-based HAR may suffer from overfitting due to the notoriously
small, often inadequate, amounts of labeled sample data that are available for
typical HAR applications. In response to such challenges, we propose ConvBoost
-- a novel, three-layer, structured model architecture and boosting framework
for convolutional network based HAR. Our framework generates additional
training data from three different perspectives for improved HAR, aiming to
alleviate the shortness of labeled training data in the field. Specifically,
with the introduction of three conceptual layers--Sampling Layer, Data
Augmentation Layer, and Resilient Layer -- we develop three "boosters" --
R-Frame, Mix-up, and C-Drop -- to enrich the per-epoch training data by
dense-sampling, synthesizing, and simulating, respectively. These new
conceptual layers and boosters, that are universally applicable for any kind of
convolutional network, have been designed based on the characteristics of the
sensor data and the concept of frame-wise HAR. In our experimental evaluation
on three standard benchmarks (Opportunity, PAMAP2, GOTOV) we demonstrate the
effectiveness of our ConvBoost framework for HAR applications based on variants
of convolutional networks: vanilla CNN, ConvLSTM, and Attention Models. We
achieved substantial performance gains for all of them, which suggests that the
proposed approach is generic and can serve as a practical solution for boosting
the performance of existing ConvNet-based HAR models. This is an open-source
project, and the code can be found at https://github.com/sshao2013/ConvBoost
| no_new_dataset | 0.710478 |
2305.13542 | Judy Hanwen Shen | Inbal Livni Navon, Charlotte Peale, Omer Reingold, Judy Hanwen Shen | Bidding Strategies for Proportional Representation in Advertisement
Campaigns | Foundations of Responsible Computing (FORC 2023) | null | 10.4230/LIPIcs.FORC.2023.3 | null | cs.CY | http://creativecommons.org/licenses/by/4.0/ | Many companies rely on advertising platforms such as Google, Facebook, or
Instagram to recruit a large and diverse applicant pool for job openings. Prior
works have shown that equitable bidding may not result in equitable outcomes
due to heterogeneous levels of competition for different types of individuals.
Suggestions have been made to address this problem via revisions to the
advertising platform. However, it may be challenging to convince platforms to
undergo a costly re-vamp of their system, and in addition it might not offer
the flexibility necessary to capture the many types of fairness notions and
other constraints that advertisers would like to ensure. Instead, we consider
alterations that make no change to the platform mechanism and instead change
the bidding strategies used by advertisers. We compare two natural fairness
objectives: one in which the advertisers must treat groups equally when bidding
in order to achieve a yield with group-parity guarantees, and another in which
the bids are not constrained and only the yield must satisfy parity
constraints. We show that requiring parity with respect to both bids and yield
can result in an arbitrarily large decrease in efficiency compared to requiring
equal yield proportions alone. We find that autobidding is a natural way to
realize this latter objective and show how existing work in this area can be
extended to provide efficient bidding strategies that provide high utility
while satisfying group parity constraints as well as deterministic and
randomized rounding techniques to uphold these guarantees. Finally, we
demonstrate the effectiveness of our proposed solutions on data adapted from a
real-world employment dataset.
| [
{
"version": "v1",
"created": "Mon, 22 May 2023 23:29:05 GMT"
}
] | 2023-05-24T00:00:00 | [
[
"Navon",
"Inbal Livni",
""
],
[
"Peale",
"Charlotte",
""
],
[
"Reingold",
"Omer",
""
],
[
"Shen",
"Judy Hanwen",
""
]
] | TITLE: Bidding Strategies for Proportional Representation in Advertisement
Campaigns
ABSTRACT: Many companies rely on advertising platforms such as Google, Facebook, or
Instagram to recruit a large and diverse applicant pool for job openings. Prior
works have shown that equitable bidding may not result in equitable outcomes
due to heterogeneous levels of competition for different types of individuals.
Suggestions have been made to address this problem via revisions to the
advertising platform. However, it may be challenging to convince platforms to
undergo a costly re-vamp of their system, and in addition it might not offer
the flexibility necessary to capture the many types of fairness notions and
other constraints that advertisers would like to ensure. Instead, we consider
alterations that make no change to the platform mechanism and instead change
the bidding strategies used by advertisers. We compare two natural fairness
objectives: one in which the advertisers must treat groups equally when bidding
in order to achieve a yield with group-parity guarantees, and another in which
the bids are not constrained and only the yield must satisfy parity
constraints. We show that requiring parity with respect to both bids and yield
can result in an arbitrarily large decrease in efficiency compared to requiring
equal yield proportions alone. We find that autobidding is a natural way to
realize this latter objective and show how existing work in this area can be
extended to provide efficient bidding strategies that provide high utility
while satisfying group parity constraints as well as deterministic and
randomized rounding techniques to uphold these guarantees. Finally, we
demonstrate the effectiveness of our proposed solutions on data adapted from a
real-world employment dataset.
| no_new_dataset | 0.709044 |
2305.13548 | Chun Pong Lau | Chun Pong Lau, Jiang Liu, Rama Chellappa | Attribute-Guided Encryption with Facial Texture Masking | null | null | null | null | cs.CV cs.CR | http://creativecommons.org/licenses/by/4.0/ | The increasingly pervasive facial recognition (FR) systems raise serious
concerns about personal privacy, especially for billions of users who have
publicly shared their photos on social media. Several attempts have been made
to protect individuals from unauthorized FR systems utilizing adversarial
attacks to generate encrypted face images to protect users from being
identified by FR systems. However, existing methods suffer from poor visual
quality or low attack success rates, which limit their usability in practice.
In this paper, we propose Attribute Guided Encryption with Facial Texture
Masking (AGE-FTM) that performs a dual manifold adversarial attack on FR
systems to achieve both good visual quality and high black box attack success
rates. In particular, AGE-FTM utilizes a high fidelity generative adversarial
network (GAN) to generate natural on-manifold adversarial samples by modifying
facial attributes, and performs the facial texture masking attack to generate
imperceptible off-manifold adversarial samples. Extensive experiments on the
CelebA-HQ dataset demonstrate that our proposed method produces more
natural-looking encrypted images than state-of-the-art methods while achieving
competitive attack performance. We further evaluate the effectiveness of
AGE-FTM in the real world using a commercial FR API and validate its usefulness
in practice through an user study.
| [
{
"version": "v1",
"created": "Mon, 22 May 2023 23:50:43 GMT"
}
] | 2023-05-24T00:00:00 | [
[
"Lau",
"Chun Pong",
""
],
[
"Liu",
"Jiang",
""
],
[
"Chellappa",
"Rama",
""
]
] | TITLE: Attribute-Guided Encryption with Facial Texture Masking
ABSTRACT: The increasingly pervasive facial recognition (FR) systems raise serious
concerns about personal privacy, especially for billions of users who have
publicly shared their photos on social media. Several attempts have been made
to protect individuals from unauthorized FR systems utilizing adversarial
attacks to generate encrypted face images to protect users from being
identified by FR systems. However, existing methods suffer from poor visual
quality or low attack success rates, which limit their usability in practice.
In this paper, we propose Attribute Guided Encryption with Facial Texture
Masking (AGE-FTM) that performs a dual manifold adversarial attack on FR
systems to achieve both good visual quality and high black box attack success
rates. In particular, AGE-FTM utilizes a high fidelity generative adversarial
network (GAN) to generate natural on-manifold adversarial samples by modifying
facial attributes, and performs the facial texture masking attack to generate
imperceptible off-manifold adversarial samples. Extensive experiments on the
CelebA-HQ dataset demonstrate that our proposed method produces more
natural-looking encrypted images than state-of-the-art methods while achieving
competitive attack performance. We further evaluate the effectiveness of
AGE-FTM in the real world using a commercial FR API and validate its usefulness
in practice through an user study.
| no_new_dataset | 0.71145 |
2305.13573 | Jintang Li | Sheng Tian, Jihai Dong, Jintang Li, Wenlong Zhao, Xiaolong Xu, Baokun
wang, Bowen Song, Changhua Meng, Tianyi Zhang, Liang Chen | SAD: Semi-Supervised Anomaly Detection on Dynamic Graphs | Accepted to IJCAI'23. Code will be available at
https://github.com/D10Andy/SAD | null | null | null | cs.LG cs.SI | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Anomaly detection aims to distinguish abnormal instances that deviate
significantly from the majority of benign ones. As instances that appear in the
real world are naturally connected and can be represented with graphs, graph
neural networks become increasingly popular in tackling the anomaly detection
problem. Despite the promising results, research on anomaly detection has
almost exclusively focused on static graphs while the mining of anomalous
patterns from dynamic graphs is rarely studied but has significant application
value. In addition, anomaly detection is typically tackled from semi-supervised
perspectives due to the lack of sufficient labeled data. However, most proposed
methods are limited to merely exploiting labeled data, leaving a large number
of unlabeled samples unexplored. In this work, we present semi-supervised
anomaly detection (SAD), an end-to-end framework for anomaly detection on
dynamic graphs. By a combination of a time-equipped memory bank and a
pseudo-label contrastive learning module, SAD is able to fully exploit the
potential of large unlabeled samples and uncover underlying anomalies on
evolving graph streams. Extensive experiments on four real-world datasets
demonstrate that SAD efficiently discovers anomalies from dynamic graphs and
outperforms existing advanced methods even when provided with only little
labeled data.
| [
{
"version": "v1",
"created": "Tue, 23 May 2023 01:05:34 GMT"
}
] | 2023-05-24T00:00:00 | [
[
"Tian",
"Sheng",
""
],
[
"Dong",
"Jihai",
""
],
[
"Li",
"Jintang",
""
],
[
"Zhao",
"Wenlong",
""
],
[
"Xu",
"Xiaolong",
""
],
[
"wang",
"Baokun",
""
],
[
"Song",
"Bowen",
""
],
[
"Meng",
"Changhua",
""
],
[
"Zhang",
"Tianyi",
""
],
[
"Chen",
"Liang",
""
]
] | TITLE: SAD: Semi-Supervised Anomaly Detection on Dynamic Graphs
ABSTRACT: Anomaly detection aims to distinguish abnormal instances that deviate
significantly from the majority of benign ones. As instances that appear in the
real world are naturally connected and can be represented with graphs, graph
neural networks become increasingly popular in tackling the anomaly detection
problem. Despite the promising results, research on anomaly detection has
almost exclusively focused on static graphs while the mining of anomalous
patterns from dynamic graphs is rarely studied but has significant application
value. In addition, anomaly detection is typically tackled from semi-supervised
perspectives due to the lack of sufficient labeled data. However, most proposed
methods are limited to merely exploiting labeled data, leaving a large number
of unlabeled samples unexplored. In this work, we present semi-supervised
anomaly detection (SAD), an end-to-end framework for anomaly detection on
dynamic graphs. By a combination of a time-equipped memory bank and a
pseudo-label contrastive learning module, SAD is able to fully exploit the
potential of large unlabeled samples and uncover underlying anomalies on
evolving graph streams. Extensive experiments on four real-world datasets
demonstrate that SAD efficiently discovers anomalies from dynamic graphs and
outperforms existing advanced methods even when provided with only little
labeled data.
| no_new_dataset | 0.709466 |
2305.13580 | Marc Delcroix | Marc Delcroix, Naohiro Tawara, Mireia Diez, Federico Landini, Anna
Silnova, Atsunori Ogawa, Tomohiro Nakatani, Lukas Burget, Shoko Araki | Multi-Stream Extension of Variational Bayesian HMM Clustering (MS-VBx)
for Combined End-to-End and Vector Clustering-based Diarization | Accepted at Interspeech 2023 | null | null | null | eess.AS cs.SD | http://creativecommons.org/licenses/by-nc-nd/4.0/ | Combining end-to-end neural speaker diarization (EEND) with vector clustering
(VC), known as EEND-VC, has gained interest for leveraging the strengths of
both methods. EEND-VC estimates activities and speaker embeddings for all
speakers within an audio chunk and uses VC to associate these activities with
speaker identities across different chunks. EEND-VC generates thus multiple
streams of embeddings, one for each speaker in a chunk. We can cluster these
embeddings using constrained agglomerative hierarchical clustering (cAHC),
ensuring embeddings from the same chunk belong to different clusters. This
paper introduces an alternative clustering approach, a multi-stream extension
of the successful Bayesian HMM clustering of x-vectors (VBx), called MS-VBx.
Experiments on three datasets demonstrate that MS-VBx outperforms cAHC in
diarization and speaker counting performance.
| [
{
"version": "v1",
"created": "Tue, 23 May 2023 01:19:28 GMT"
}
] | 2023-05-24T00:00:00 | [
[
"Delcroix",
"Marc",
""
],
[
"Tawara",
"Naohiro",
""
],
[
"Diez",
"Mireia",
""
],
[
"Landini",
"Federico",
""
],
[
"Silnova",
"Anna",
""
],
[
"Ogawa",
"Atsunori",
""
],
[
"Nakatani",
"Tomohiro",
""
],
[
"Burget",
"Lukas",
""
],
[
"Araki",
"Shoko",
""
]
] | TITLE: Multi-Stream Extension of Variational Bayesian HMM Clustering (MS-VBx)
for Combined End-to-End and Vector Clustering-based Diarization
ABSTRACT: Combining end-to-end neural speaker diarization (EEND) with vector clustering
(VC), known as EEND-VC, has gained interest for leveraging the strengths of
both methods. EEND-VC estimates activities and speaker embeddings for all
speakers within an audio chunk and uses VC to associate these activities with
speaker identities across different chunks. EEND-VC generates thus multiple
streams of embeddings, one for each speaker in a chunk. We can cluster these
embeddings using constrained agglomerative hierarchical clustering (cAHC),
ensuring embeddings from the same chunk belong to different clusters. This
paper introduces an alternative clustering approach, a multi-stream extension
of the successful Bayesian HMM clustering of x-vectors (VBx), called MS-VBx.
Experiments on three datasets demonstrate that MS-VBx outperforms cAHC in
diarization and speaker counting performance.
| no_new_dataset | 0.710465 |
2305.13585 | Siyuan Wang | Siyuan Wang, Zhongyu Wei, Meng Han, Zhihao Fan, Haijun Shan, Qi Zhang,
Xuanjing Huang | Query Structure Modeling for Inductive Logical Reasoning Over Knowledge
Graphs | 11 pages, 2 figures, 8 tables, accepted as a long paper to ACL 203 | null | null | null | cs.CL | http://creativecommons.org/licenses/by/4.0/ | Logical reasoning over incomplete knowledge graphs to answer complex logical
queries is a challenging task. With the emergence of new entities and relations
in constantly evolving KGs, inductive logical reasoning over KGs has become a
crucial problem. However, previous PLMs-based methods struggle to model the
logical structures of complex queries, which limits their ability to generalize
within the same structure. In this paper, we propose a structure-modeled
textual encoding framework for inductive logical reasoning over KGs. It encodes
linearized query structures and entities using pre-trained language models to
find answers. For structure modeling of complex queries, we design stepwise
instructions that implicitly prompt PLMs on the execution order of geometric
operations in each query. We further separately model different geometric
operations (i.e., projection, intersection, and union) on the representation
space using a pre-trained encoder with additional attention and maxout layers
to enhance structured modeling. We conduct experiments on two inductive logical
reasoning datasets and three transductive datasets. The results demonstrate the
effectiveness of our method on logical reasoning over KGs in both inductive and
transductive settings.
| [
{
"version": "v1",
"created": "Tue, 23 May 2023 01:25:29 GMT"
}
] | 2023-05-24T00:00:00 | [
[
"Wang",
"Siyuan",
""
],
[
"Wei",
"Zhongyu",
""
],
[
"Han",
"Meng",
""
],
[
"Fan",
"Zhihao",
""
],
[
"Shan",
"Haijun",
""
],
[
"Zhang",
"Qi",
""
],
[
"Huang",
"Xuanjing",
""
]
] | TITLE: Query Structure Modeling for Inductive Logical Reasoning Over Knowledge
Graphs
ABSTRACT: Logical reasoning over incomplete knowledge graphs to answer complex logical
queries is a challenging task. With the emergence of new entities and relations
in constantly evolving KGs, inductive logical reasoning over KGs has become a
crucial problem. However, previous PLMs-based methods struggle to model the
logical structures of complex queries, which limits their ability to generalize
within the same structure. In this paper, we propose a structure-modeled
textual encoding framework for inductive logical reasoning over KGs. It encodes
linearized query structures and entities using pre-trained language models to
find answers. For structure modeling of complex queries, we design stepwise
instructions that implicitly prompt PLMs on the execution order of geometric
operations in each query. We further separately model different geometric
operations (i.e., projection, intersection, and union) on the representation
space using a pre-trained encoder with additional attention and maxout layers
to enhance structured modeling. We conduct experiments on two inductive logical
reasoning datasets and three transductive datasets. The results demonstrate the
effectiveness of our method on logical reasoning over KGs in both inductive and
transductive settings.
| no_new_dataset | 0.709044 |
2305.13591 | Mingshuai Dong | Mingshuai Dong and Yuxuan Bai and Shimin Wei and Xiuli Yu | A Single Multi-Task Deep Neural Network with a Multi-Scale Feature
Aggregation Mechanism for Manipulation Relationship Reasoning in Robotic
Grasping | null | null | null | null | cs.RO | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Grasping specific objects in complex and irregularly stacked scenes is still
challenging for robotics. Because the robot is not only required to identify
the object's grasping posture but also needs to reason the manipulation
relationship between the objects. In this paper, we propose a manipulation
relationship reasoning network with a multi-scale feature aggregation (MSFA)
mechanism for robot grasping tasks. MSFA aggregates high-level semantic
information and low-level spatial information in a cross-scale connection way
to improve the generalization ability of the model. Furthermore, to improve the
accuracy, we propose to use intersection features with rich location priors for
manipulation relationship reasoning. Experiments are validated in VMRD datasets
and real environments, respectively. The experimental results demonstrate that
our proposed method can accurately predict the manipulation relationship
between objects in the scene of multi-object stacking. Compared with previous
methods, it significantly improves reasoning speed and accuracy.
| [
{
"version": "v1",
"created": "Tue, 23 May 2023 01:46:25 GMT"
}
] | 2023-05-24T00:00:00 | [
[
"Dong",
"Mingshuai",
""
],
[
"Bai",
"Yuxuan",
""
],
[
"Wei",
"Shimin",
""
],
[
"Yu",
"Xiuli",
""
]
] | TITLE: A Single Multi-Task Deep Neural Network with a Multi-Scale Feature
Aggregation Mechanism for Manipulation Relationship Reasoning in Robotic
Grasping
ABSTRACT: Grasping specific objects in complex and irregularly stacked scenes is still
challenging for robotics. Because the robot is not only required to identify
the object's grasping posture but also needs to reason the manipulation
relationship between the objects. In this paper, we propose a manipulation
relationship reasoning network with a multi-scale feature aggregation (MSFA)
mechanism for robot grasping tasks. MSFA aggregates high-level semantic
information and low-level spatial information in a cross-scale connection way
to improve the generalization ability of the model. Furthermore, to improve the
accuracy, we propose to use intersection features with rich location priors for
manipulation relationship reasoning. Experiments are validated in VMRD datasets
and real environments, respectively. The experimental results demonstrate that
our proposed method can accurately predict the manipulation relationship
between objects in the scene of multi-object stacking. Compared with previous
methods, it significantly improves reasoning speed and accuracy.
| no_new_dataset | 0.711588 |
2305.13605 | Mei Wang | Mei Wang, Weihong Deng | Adaptive Face Recognition Using Adversarial Information Network | Accepted by TIP | null | null | null | cs.CV | http://creativecommons.org/licenses/by/4.0/ | In many real-world applications, face recognition models often degenerate
when training data (referred to as source domain) are different from testing
data (referred to as target domain). To alleviate this mismatch caused by some
factors like pose and skin tone, the utilization of pseudo-labels generated by
clustering algorithms is an effective way in unsupervised domain adaptation.
However, they always miss some hard positive samples. Supervision on
pseudo-labeled samples attracts them towards their prototypes and would cause
an intra-domain gap between pseudo-labeled samples and the remaining unlabeled
samples within target domain, which results in the lack of discrimination in
face recognition. In this paper, considering the particularity of face
recognition, we propose a novel adversarial information network (AIN) to
address it. First, a novel adversarial mutual information (MI) loss is proposed
to alternately minimize MI with respect to the target classifier and maximize
MI with respect to the feature extractor. By this min-max manner, the positions
of target prototypes are adaptively modified which makes unlabeled images
clustered more easily such that intra-domain gap can be mitigated. Second, to
assist adversarial MI loss, we utilize a graph convolution network to predict
linkage likelihoods between target data and generate pseudo-labels. It
leverages valuable information in the context of nodes and can achieve more
reliable results. The proposed method is evaluated under two scenarios, i.e.,
domain adaptation across poses and image conditions, and domain adaptation
across faces with different skin tones. Extensive experiments show that AIN
successfully improves cross-domain generalization and offers a new
state-of-the-art on RFW dataset.
| [
{
"version": "v1",
"created": "Tue, 23 May 2023 02:14:11 GMT"
}
] | 2023-05-24T00:00:00 | [
[
"Wang",
"Mei",
""
],
[
"Deng",
"Weihong",
""
]
] | TITLE: Adaptive Face Recognition Using Adversarial Information Network
ABSTRACT: In many real-world applications, face recognition models often degenerate
when training data (referred to as source domain) are different from testing
data (referred to as target domain). To alleviate this mismatch caused by some
factors like pose and skin tone, the utilization of pseudo-labels generated by
clustering algorithms is an effective way in unsupervised domain adaptation.
However, they always miss some hard positive samples. Supervision on
pseudo-labeled samples attracts them towards their prototypes and would cause
an intra-domain gap between pseudo-labeled samples and the remaining unlabeled
samples within target domain, which results in the lack of discrimination in
face recognition. In this paper, considering the particularity of face
recognition, we propose a novel adversarial information network (AIN) to
address it. First, a novel adversarial mutual information (MI) loss is proposed
to alternately minimize MI with respect to the target classifier and maximize
MI with respect to the feature extractor. By this min-max manner, the positions
of target prototypes are adaptively modified which makes unlabeled images
clustered more easily such that intra-domain gap can be mitigated. Second, to
assist adversarial MI loss, we utilize a graph convolution network to predict
linkage likelihoods between target data and generate pseudo-labels. It
leverages valuable information in the context of nodes and can achieve more
reliable results. The proposed method is evaluated under two scenarios, i.e.,
domain adaptation across poses and image conditions, and domain adaptation
across faces with different skin tones. Extensive experiments show that AIN
successfully improves cross-domain generalization and offers a new
state-of-the-art on RFW dataset.
| no_new_dataset | 0.711606 |
2305.13611 | Yue Lu | Congqi Cao, Yue Lu, Peng Wang and Yanning Zhang | A New Comprehensive Benchmark for Semi-supervised Video Anomaly
Detection and Anticipation | CVPR 2023 | null | null | null | cs.CV | http://creativecommons.org/licenses/by-nc-nd/4.0/ | Semi-supervised video anomaly detection (VAD) is a critical task in the
intelligent surveillance system. However, an essential type of anomaly in VAD
named scene-dependent anomaly has not received the attention of researchers.
Moreover, there is no research investigating anomaly anticipation, a more
significant task for preventing the occurrence of anomalous events. To this
end, we propose a new comprehensive dataset, NWPU Campus, containing 43 scenes,
28 classes of abnormal events, and 16 hours of videos. At present, it is the
largest semi-supervised VAD dataset with the largest number of scenes and
classes of anomalies, the longest duration, and the only one considering the
scene-dependent anomaly. Meanwhile, it is also the first dataset proposed for
video anomaly anticipation. We further propose a novel model capable of
detecting and anticipating anomalous events simultaneously. Compared with 7
outstanding VAD algorithms in recent years, our method can cope with
scene-dependent anomaly detection and anomaly anticipation both well, achieving
state-of-the-art performance on ShanghaiTech, CUHK Avenue, IITB Corridor and
the newly proposed NWPU Campus datasets consistently. Our dataset and code is
available at: https://campusvad.github.io.
| [
{
"version": "v1",
"created": "Tue, 23 May 2023 02:20:12 GMT"
}
] | 2023-05-24T00:00:00 | [
[
"Cao",
"Congqi",
""
],
[
"Lu",
"Yue",
""
],
[
"Wang",
"Peng",
""
],
[
"Zhang",
"Yanning",
""
]
] | TITLE: A New Comprehensive Benchmark for Semi-supervised Video Anomaly
Detection and Anticipation
ABSTRACT: Semi-supervised video anomaly detection (VAD) is a critical task in the
intelligent surveillance system. However, an essential type of anomaly in VAD
named scene-dependent anomaly has not received the attention of researchers.
Moreover, there is no research investigating anomaly anticipation, a more
significant task for preventing the occurrence of anomalous events. To this
end, we propose a new comprehensive dataset, NWPU Campus, containing 43 scenes,
28 classes of abnormal events, and 16 hours of videos. At present, it is the
largest semi-supervised VAD dataset with the largest number of scenes and
classes of anomalies, the longest duration, and the only one considering the
scene-dependent anomaly. Meanwhile, it is also the first dataset proposed for
video anomaly anticipation. We further propose a novel model capable of
detecting and anticipating anomalous events simultaneously. Compared with 7
outstanding VAD algorithms in recent years, our method can cope with
scene-dependent anomaly detection and anomaly anticipation both well, achieving
state-of-the-art performance on ShanghaiTech, CUHK Avenue, IITB Corridor and
the newly proposed NWPU Campus datasets consistently. Our dataset and code is
available at: https://campusvad.github.io.
| new_dataset | 0.713619 |
2305.13612 | Ziyue Jiang | Ziyue Jiang, Qian Yang, Jialong Zuo, Zhenhui Ye, Rongjie Huang, Yi
Ren, Zhou Zhao | FluentSpeech: Stutter-Oriented Automatic Speech Editing with
Context-Aware Diffusion Models | Accepted by ACL 2023 (Findings) | null | null | null | cs.SD eess.AS | http://creativecommons.org/publicdomain/zero/1.0/ | Stutter removal is an essential scenario in the field of speech editing.
However, when the speech recording contains stutters, the existing text-based
speech editing approaches still suffer from: 1) the over-smoothing problem in
the edited speech; 2) lack of robustness due to the noise introduced by
stutter; 3) to remove the stutters, users are required to determine the edited
region manually. To tackle the challenges in stutter removal, we propose
FluentSpeech, a stutter-oriented automatic speech editing model. Specifically,
1) we propose a context-aware diffusion model that iteratively refines the
modified mel-spectrogram with the guidance of context features; 2) we introduce
a stutter predictor module to inject the stutter information into the hidden
sequence; 3) we also propose a stutter-oriented automatic speech editing (SASE)
dataset that contains spontaneous speech recordings with time-aligned stutter
labels to train the automatic stutter localization model. Experimental results
on VCTK and LibriTTS datasets demonstrate that our model achieves
state-of-the-art performance on speech editing. Further experiments on our SASE
dataset show that FluentSpeech can effectively improve the fluency of
stuttering speech in terms of objective and subjective metrics. Code and audio
samples can be found at https://github.com/Zain-Jiang/Speech-Editing-Toolkit.
| [
{
"version": "v1",
"created": "Tue, 23 May 2023 02:20:47 GMT"
}
] | 2023-05-24T00:00:00 | [
[
"Jiang",
"Ziyue",
""
],
[
"Yang",
"Qian",
""
],
[
"Zuo",
"Jialong",
""
],
[
"Ye",
"Zhenhui",
""
],
[
"Huang",
"Rongjie",
""
],
[
"Ren",
"Yi",
""
],
[
"Zhao",
"Zhou",
""
]
] | TITLE: FluentSpeech: Stutter-Oriented Automatic Speech Editing with
Context-Aware Diffusion Models
ABSTRACT: Stutter removal is an essential scenario in the field of speech editing.
However, when the speech recording contains stutters, the existing text-based
speech editing approaches still suffer from: 1) the over-smoothing problem in
the edited speech; 2) lack of robustness due to the noise introduced by
stutter; 3) to remove the stutters, users are required to determine the edited
region manually. To tackle the challenges in stutter removal, we propose
FluentSpeech, a stutter-oriented automatic speech editing model. Specifically,
1) we propose a context-aware diffusion model that iteratively refines the
modified mel-spectrogram with the guidance of context features; 2) we introduce
a stutter predictor module to inject the stutter information into the hidden
sequence; 3) we also propose a stutter-oriented automatic speech editing (SASE)
dataset that contains spontaneous speech recordings with time-aligned stutter
labels to train the automatic stutter localization model. Experimental results
on VCTK and LibriTTS datasets demonstrate that our model achieves
state-of-the-art performance on speech editing. Further experiments on our SASE
dataset show that FluentSpeech can effectively improve the fluency of
stuttering speech in terms of objective and subjective metrics. Code and audio
samples can be found at https://github.com/Zain-Jiang/Speech-Editing-Toolkit.
| no_new_dataset | 0.589126 |
2305.13634 | Zekun Qiu | Zekun Qiu, Zhipu Xie, Zehua Ji, Yuhao Mao, Ke Cheng | SMAP: A Novel Heterogeneous Information Framework for Scenario-based
Optimal Model Assignment | null | null | null | null | cs.LG cs.AI | http://creativecommons.org/licenses/by-nc-sa/4.0/ | The increasing maturity of big data applications has led to a proliferation
of models targeting the same objectives within the same scenarios and datasets.
However, selecting the most suitable model that considers model's features
while taking specific requirements and constraints into account still poses a
significant challenge. Existing methods have focused on worker-task assignments
based on crowdsourcing, they neglect the scenario-dataset-model assignment
problem. To address this challenge, a new problem named the Scenario-based
Optimal Model Assignment (SOMA) problem is introduced and a novel framework
entitled Scenario and Model Associative percepts (SMAP) is developed. SMAP is a
heterogeneous information framework that can integrate various types of
information to intelligently select a suitable dataset and allocate the optimal
model for a specific scenario. To comprehensively evaluate models, a new score
function that utilizes multi-head attention mechanisms is proposed. Moreover, a
novel memory mechanism named the mnemonic center is developed to store the
matched heterogeneous information and prevent duplicate matching. Six popular
traffic scenarios are selected as study cases and extensive experiments are
conducted on a dataset to verify the effectiveness and efficiency of SMAP and
the score function.
| [
{
"version": "v1",
"created": "Tue, 23 May 2023 03:01:26 GMT"
}
] | 2023-05-24T00:00:00 | [
[
"Qiu",
"Zekun",
""
],
[
"Xie",
"Zhipu",
""
],
[
"Ji",
"Zehua",
""
],
[
"Mao",
"Yuhao",
""
],
[
"Cheng",
"Ke",
""
]
] | TITLE: SMAP: A Novel Heterogeneous Information Framework for Scenario-based
Optimal Model Assignment
ABSTRACT: The increasing maturity of big data applications has led to a proliferation
of models targeting the same objectives within the same scenarios and datasets.
However, selecting the most suitable model that considers model's features
while taking specific requirements and constraints into account still poses a
significant challenge. Existing methods have focused on worker-task assignments
based on crowdsourcing, they neglect the scenario-dataset-model assignment
problem. To address this challenge, a new problem named the Scenario-based
Optimal Model Assignment (SOMA) problem is introduced and a novel framework
entitled Scenario and Model Associative percepts (SMAP) is developed. SMAP is a
heterogeneous information framework that can integrate various types of
information to intelligently select a suitable dataset and allocate the optimal
model for a specific scenario. To comprehensively evaluate models, a new score
function that utilizes multi-head attention mechanisms is proposed. Moreover, a
novel memory mechanism named the mnemonic center is developed to store the
matched heterogeneous information and prevent duplicate matching. Six popular
traffic scenarios are selected as study cases and extensive experiments are
conducted on a dataset to verify the effectiveness and efficiency of SMAP and
the score function.
| no_new_dataset | 0.709447 |
2305.13648 | Jiayi Wang | Jiayi Wang, Ke Wang, Yuqi Zhang, Yu Zhao, Pontus Stenetorp | Non-parametric, Nearest-neighbor-assisted Fine-tuning for Neural Machine
Translation | null | null | null | null | cs.CL cs.AI | http://creativecommons.org/licenses/by-nc-nd/4.0/ | Non-parametric, k-nearest-neighbor algorithms have recently made inroads to
assist generative models such as language models and machine translation
decoders. We explore whether such non-parametric models can improve machine
translation models at the fine-tuning stage by incorporating statistics from
the kNN predictions to inform the gradient updates for a baseline translation
model. There are multiple methods which could be used to incorporate kNN
statistics and we investigate gradient scaling by a gating mechanism, the kNN's
ground truth probability, and reinforcement learning. For four standard
in-domain machine translation datasets, compared with classic fine-tuning, we
report consistent improvements of all of the three methods by as much as 1.45
BLEU and 1.28 BLEU for German-English and English-German translations
respectively. Through qualitative analysis, we found particular improvements
when it comes to translating grammatical relations or function words, which
results in increased fluency of our model.
| [
{
"version": "v1",
"created": "Tue, 23 May 2023 03:44:06 GMT"
}
] | 2023-05-24T00:00:00 | [
[
"Wang",
"Jiayi",
""
],
[
"Wang",
"Ke",
""
],
[
"Zhang",
"Yuqi",
""
],
[
"Zhao",
"Yu",
""
],
[
"Stenetorp",
"Pontus",
""
]
] | TITLE: Non-parametric, Nearest-neighbor-assisted Fine-tuning for Neural Machine
Translation
ABSTRACT: Non-parametric, k-nearest-neighbor algorithms have recently made inroads to
assist generative models such as language models and machine translation
decoders. We explore whether such non-parametric models can improve machine
translation models at the fine-tuning stage by incorporating statistics from
the kNN predictions to inform the gradient updates for a baseline translation
model. There are multiple methods which could be used to incorporate kNN
statistics and we investigate gradient scaling by a gating mechanism, the kNN's
ground truth probability, and reinforcement learning. For four standard
in-domain machine translation datasets, compared with classic fine-tuning, we
report consistent improvements of all of the three methods by as much as 1.45
BLEU and 1.28 BLEU for German-English and English-German translations
respectively. Through qualitative analysis, we found particular improvements
when it comes to translating grammatical relations or function words, which
results in increased fluency of our model.
| no_new_dataset | 0.710208 |
2305.13663 | Shuai Wang | Lexiong Huang, Ruihua Han, Guoliang Li, He Li, Shuai Wang, Yang Wang,
Chengzhong Xu | iCOIL: Scenario Aware Autonomous Parking Via Integrated Constrained
Optimization and Imitation Learning | 6 pages, 8 figures, IEEE ICDCS Workshops, 2023 | null | null | null | cs.RO | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Autonomous parking (AP) is an emering technique to navigate an intelligent
vehicle to a parking space without any human intervention. Existing AP methods
based on mathematical optimization or machine learning may lead to potential
failures due to either excessive execution time or lack of generalization. To
fill this gap, this paper proposes an integrated constrained optimization and
imitation learning (iCOIL) approach to achieve efficient and reliable AP. The
iCOIL method has two candidate working modes, i.e., CO and IL, and adopts a
hybrid scenario analysis (HSA) model to determine the better mode under various
scenarios. We implement and verify iCOIL on the Macao Car Racing Metaverse
(MoCAM) platform. Results show that iCOIL properly adapts to different
scenarios during the entire AP procedure, and achieves significantly larger
success rates than other benchmarks.
| [
{
"version": "v1",
"created": "Tue, 23 May 2023 04:12:52 GMT"
}
] | 2023-05-24T00:00:00 | [
[
"Huang",
"Lexiong",
""
],
[
"Han",
"Ruihua",
""
],
[
"Li",
"Guoliang",
""
],
[
"Li",
"He",
""
],
[
"Wang",
"Shuai",
""
],
[
"Wang",
"Yang",
""
],
[
"Xu",
"Chengzhong",
""
]
] | TITLE: iCOIL: Scenario Aware Autonomous Parking Via Integrated Constrained
Optimization and Imitation Learning
ABSTRACT: Autonomous parking (AP) is an emering technique to navigate an intelligent
vehicle to a parking space without any human intervention. Existing AP methods
based on mathematical optimization or machine learning may lead to potential
failures due to either excessive execution time or lack of generalization. To
fill this gap, this paper proposes an integrated constrained optimization and
imitation learning (iCOIL) approach to achieve efficient and reliable AP. The
iCOIL method has two candidate working modes, i.e., CO and IL, and adopts a
hybrid scenario analysis (HSA) model to determine the better mode under various
scenarios. We implement and verify iCOIL on the Macao Car Racing Metaverse
(MoCAM) platform. Results show that iCOIL properly adapts to different
scenarios during the entire AP procedure, and achieves significantly larger
success rates than other benchmarks.
| no_new_dataset | 0.709617 |
2305.13665 | Linwei Tao | Linwei Tao, Minjing Dong and Chang Xu | Dual Focal Loss for Calibration | ICML 2023 Accept | null | null | null | cs.CV cs.AI | http://creativecommons.org/licenses/by/4.0/ | The use of deep neural networks in real-world applications require
well-calibrated networks with confidence scores that accurately reflect the
actual probability. However, it has been found that these networks often
provide over-confident predictions, which leads to poor calibration. Recent
efforts have sought to address this issue by focal loss to reduce
over-confidence, but this approach can also lead to under-confident
predictions. While different variants of focal loss have been explored, it is
difficult to find a balance between over-confidence and under-confidence. In
our work, we propose a new loss function by focusing on dual logits. Our method
not only considers the ground truth logit, but also take into account the
highest logit ranked after the ground truth logit. By maximizing the gap
between these two logits, our proposed dual focal loss can achieve a better
balance between over-confidence and under-confidence. We provide theoretical
evidence to support our approach and demonstrate its effectiveness through
evaluations on multiple models and datasets, where it achieves state-of-the-art
performance. Code is available at https://github.com/Linwei94/DualFocalLoss
| [
{
"version": "v1",
"created": "Tue, 23 May 2023 04:19:16 GMT"
}
] | 2023-05-24T00:00:00 | [
[
"Tao",
"Linwei",
""
],
[
"Dong",
"Minjing",
""
],
[
"Xu",
"Chang",
""
]
] | TITLE: Dual Focal Loss for Calibration
ABSTRACT: The use of deep neural networks in real-world applications require
well-calibrated networks with confidence scores that accurately reflect the
actual probability. However, it has been found that these networks often
provide over-confident predictions, which leads to poor calibration. Recent
efforts have sought to address this issue by focal loss to reduce
over-confidence, but this approach can also lead to under-confident
predictions. While different variants of focal loss have been explored, it is
difficult to find a balance between over-confidence and under-confidence. In
our work, we propose a new loss function by focusing on dual logits. Our method
not only considers the ground truth logit, but also take into account the
highest logit ranked after the ground truth logit. By maximizing the gap
between these two logits, our proposed dual focal loss can achieve a better
balance between over-confidence and under-confidence. We provide theoretical
evidence to support our approach and demonstrate its effectiveness through
evaluations on multiple models and datasets, where it achieves state-of-the-art
performance. Code is available at https://github.com/Linwei94/DualFocalLoss
| no_new_dataset | 0.710226 |
2305.13678 | Minchan Kwon | Minchan Kwon, Kangil Kim | Enhancing Accuracy and Robustness through Adversarial Training in Class
Incremental Continual Learning | 9 pages, 6 figures | null | null | null | cs.LG | http://creativecommons.org/licenses/by/4.0/ | In real life, adversarial attack to deep learning models is a fatal security
issue. However, the issue has been rarely discussed in a widely used
class-incremental continual learning (CICL). In this paper, we address problems
of applying adversarial training to CICL, which is well-known defense method
against adversarial attack. A well-known problem of CICL is class-imbalance
that biases a model to the current task by a few samples of previous tasks.
Meeting with the adversarial training, the imbalance causes another imbalance
of attack trials over tasks. Lacking clean data of a minority class by the
class-imbalance and increasing of attack trials from a majority class by the
secondary imbalance, adversarial training distorts optimal decision boundaries.
The distortion eventually decreases both accuracy and robustness than
adversarial training. To exclude the effects, we propose a straightforward but
significantly effective method, External Adversarial Training (EAT) which can
be applied to methods using experience replay. This method conduct adversarial
training to an auxiliary external model for the current task data at each time
step, and applies generated adversarial examples to train the target model. We
verify the effects on a toy problem and show significance on CICL benchmarks of
image classification. We expect that the results will be used as the first
baseline for robustness research of CICL.
| [
{
"version": "v1",
"created": "Tue, 23 May 2023 04:37:18 GMT"
}
] | 2023-05-24T00:00:00 | [
[
"Kwon",
"Minchan",
""
],
[
"Kim",
"Kangil",
""
]
] | TITLE: Enhancing Accuracy and Robustness through Adversarial Training in Class
Incremental Continual Learning
ABSTRACT: In real life, adversarial attack to deep learning models is a fatal security
issue. However, the issue has been rarely discussed in a widely used
class-incremental continual learning (CICL). In this paper, we address problems
of applying adversarial training to CICL, which is well-known defense method
against adversarial attack. A well-known problem of CICL is class-imbalance
that biases a model to the current task by a few samples of previous tasks.
Meeting with the adversarial training, the imbalance causes another imbalance
of attack trials over tasks. Lacking clean data of a minority class by the
class-imbalance and increasing of attack trials from a majority class by the
secondary imbalance, adversarial training distorts optimal decision boundaries.
The distortion eventually decreases both accuracy and robustness than
adversarial training. To exclude the effects, we propose a straightforward but
significantly effective method, External Adversarial Training (EAT) which can
be applied to methods using experience replay. This method conduct adversarial
training to an auxiliary external model for the current task data at each time
step, and applies generated adversarial examples to train the target model. We
verify the effects on a toy problem and show significance on CICL benchmarks of
image classification. We expect that the results will be used as the first
baseline for robustness research of CICL.
| no_new_dataset | 0.710415 |
2305.13685 | Jiachang Liu | Jiachang Liu, Qi Zhang, Chongyang Shi, Usman Naseem, Shoujin Wang,
Ivor Tsang | Causal Intervention for Abstractive Related Work Generation | null | null | null | null | cs.CL | http://creativecommons.org/licenses/by/4.0/ | Abstractive related work generation has attracted increasing attention in
generating coherent related work that better helps readers grasp the background
in the current research. However, most existing abstractive models ignore the
inherent causality of related work generation, leading to low quality of
generated related work and spurious correlations that affect the models'
generalizability. In this study, we argue that causal intervention can address
these limitations and improve the quality and coherence of the generated
related works. To this end, we propose a novel Causal Intervention Module for
Related Work Generation (CaM) to effectively capture causalities in the
generation process and improve the quality and coherence of the generated
related works. Specifically, we first model the relations among sentence order,
document relation, and transitional content in related work generation using a
causal graph. Then, to implement the causal intervention and mitigate the
negative impact of spurious correlations, we use do-calculus to derive ordinary
conditional probabilities and identify causal effects through CaM. Finally, we
subtly fuse CaM with Transformer to obtain an end-to-end generation model.
Extensive experiments on two real-world datasets show that causal interventions
in CaM can effectively promote the model to learn causal relations and produce
related work of higher quality and coherence.
| [
{
"version": "v1",
"created": "Tue, 23 May 2023 04:48:30 GMT"
}
] | 2023-05-24T00:00:00 | [
[
"Liu",
"Jiachang",
""
],
[
"Zhang",
"Qi",
""
],
[
"Shi",
"Chongyang",
""
],
[
"Naseem",
"Usman",
""
],
[
"Wang",
"Shoujin",
""
],
[
"Tsang",
"Ivor",
""
]
] | TITLE: Causal Intervention for Abstractive Related Work Generation
ABSTRACT: Abstractive related work generation has attracted increasing attention in
generating coherent related work that better helps readers grasp the background
in the current research. However, most existing abstractive models ignore the
inherent causality of related work generation, leading to low quality of
generated related work and spurious correlations that affect the models'
generalizability. In this study, we argue that causal intervention can address
these limitations and improve the quality and coherence of the generated
related works. To this end, we propose a novel Causal Intervention Module for
Related Work Generation (CaM) to effectively capture causalities in the
generation process and improve the quality and coherence of the generated
related works. Specifically, we first model the relations among sentence order,
document relation, and transitional content in related work generation using a
causal graph. Then, to implement the causal intervention and mitigate the
negative impact of spurious correlations, we use do-calculus to derive ordinary
conditional probabilities and identify causal effects through CaM. Finally, we
subtly fuse CaM with Transformer to obtain an end-to-end generation model.
Extensive experiments on two real-world datasets show that causal interventions
in CaM can effectively promote the model to learn causal relations and produce
related work of higher quality and coherence.
| no_new_dataset | 0.710025 |
2305.13690 | Yue Feng | Yue Feng, Hossein A. Rahmani, Aldo Lipani, Emine Yilmaz | Towards Asking Clarification Questions for Information Seeking on
Task-Oriented Dialogues | null | null | null | null | cs.CL cs.IR | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Task-oriented dialogue systems aim at providing users with task-specific
services. Users of such systems often do not know all the information about the
task they are trying to accomplish, requiring them to seek information about
the task. To provide accurate and personalized task-oriented information
seeking results, task-oriented dialogue systems need to address two potential
issues: 1) users' inability to describe their complex information needs in
their requests; and 2) ambiguous/missing information the system has about the
users. In this paper, we propose a new Multi-Attention Seq2Seq Network, named
MAS2S, which can ask questions to clarify the user's information needs and the
user's profile in task-oriented information seeking. We also extend an existing
dataset for task-oriented information seeking, leading to the \ourdataset which
contains about 100k task-oriented information seeking dialogues that are made
publicly available\footnote{Dataset and code is available at
\href{https://github.com/sweetalyssum/clarit}{https://github.com/sweetalyssum/clarit}.}.
Experimental results on \ourdataset show that MAS2S outperforms baselines on
both clarification question generation and answer prediction.
| [
{
"version": "v1",
"created": "Tue, 23 May 2023 04:56:04 GMT"
}
] | 2023-05-24T00:00:00 | [
[
"Feng",
"Yue",
""
],
[
"Rahmani",
"Hossein A.",
""
],
[
"Lipani",
"Aldo",
""
],
[
"Yilmaz",
"Emine",
""
]
] | TITLE: Towards Asking Clarification Questions for Information Seeking on
Task-Oriented Dialogues
ABSTRACT: Task-oriented dialogue systems aim at providing users with task-specific
services. Users of such systems often do not know all the information about the
task they are trying to accomplish, requiring them to seek information about
the task. To provide accurate and personalized task-oriented information
seeking results, task-oriented dialogue systems need to address two potential
issues: 1) users' inability to describe their complex information needs in
their requests; and 2) ambiguous/missing information the system has about the
users. In this paper, we propose a new Multi-Attention Seq2Seq Network, named
MAS2S, which can ask questions to clarify the user's information needs and the
user's profile in task-oriented information seeking. We also extend an existing
dataset for task-oriented information seeking, leading to the \ourdataset which
contains about 100k task-oriented information seeking dialogues that are made
publicly available\footnote{Dataset and code is available at
\href{https://github.com/sweetalyssum/clarit}{https://github.com/sweetalyssum/clarit}.}.
Experimental results on \ourdataset show that MAS2S outperforms baselines on
both clarification question generation and answer prediction.
| new_dataset | 0.717173 |
2305.13693 | Lucy Lu Wang | Lucy Lu Wang, Yulia Otmakhova, Jay DeYoung, Thinh Hung Truong, Bailey
E. Kuehl, Erin Bransom, Byron C. Wallace | Automated Metrics for Medical Multi-Document Summarization Disagree with
Human Evaluations | ACL 2023; Github: https://github.com/allenai/mslr-annotated-dataset | null | null | null | cs.CL | http://creativecommons.org/licenses/by/4.0/ | Evaluating multi-document summarization (MDS) quality is difficult. This is
especially true in the case of MDS for biomedical literature reviews, where
models must synthesize contradicting evidence reported across different
documents. Prior work has shown that rather than performing the task, models
may exploit shortcuts that are difficult to detect using standard n-gram
similarity metrics such as ROUGE. Better automated evaluation metrics are
needed, but few resources exist to assess metrics when they are proposed.
Therefore, we introduce a dataset of human-assessed summary quality facets and
pairwise preferences to encourage and support the development of better
automated evaluation methods for literature review MDS. We take advantage of
community submissions to the Multi-document Summarization for Literature Review
(MSLR) shared task to compile a diverse and representative sample of generated
summaries. We analyze how automated summarization evaluation metrics correlate
with lexical features of generated summaries, to other automated metrics
including several we propose in this work, and to aspects of human-assessed
summary quality. We find that not only do automated metrics fail to capture
aspects of quality as assessed by humans, in many cases the system rankings
produced by these metrics are anti-correlated with rankings according to human
annotators.
| [
{
"version": "v1",
"created": "Tue, 23 May 2023 05:00:59 GMT"
}
] | 2023-05-24T00:00:00 | [
[
"Wang",
"Lucy Lu",
""
],
[
"Otmakhova",
"Yulia",
""
],
[
"DeYoung",
"Jay",
""
],
[
"Truong",
"Thinh Hung",
""
],
[
"Kuehl",
"Bailey E.",
""
],
[
"Bransom",
"Erin",
""
],
[
"Wallace",
"Byron C.",
""
]
] | TITLE: Automated Metrics for Medical Multi-Document Summarization Disagree with
Human Evaluations
ABSTRACT: Evaluating multi-document summarization (MDS) quality is difficult. This is
especially true in the case of MDS for biomedical literature reviews, where
models must synthesize contradicting evidence reported across different
documents. Prior work has shown that rather than performing the task, models
may exploit shortcuts that are difficult to detect using standard n-gram
similarity metrics such as ROUGE. Better automated evaluation metrics are
needed, but few resources exist to assess metrics when they are proposed.
Therefore, we introduce a dataset of human-assessed summary quality facets and
pairwise preferences to encourage and support the development of better
automated evaluation methods for literature review MDS. We take advantage of
community submissions to the Multi-document Summarization for Literature Review
(MSLR) shared task to compile a diverse and representative sample of generated
summaries. We analyze how automated summarization evaluation metrics correlate
with lexical features of generated summaries, to other automated metrics
including several we propose in this work, and to aspects of human-assessed
summary quality. We find that not only do automated metrics fail to capture
aspects of quality as assessed by humans, in many cases the system rankings
produced by these metrics are anti-correlated with rankings according to human
annotators.
| new_dataset | 0.714983 |
2305.13698 | Frederick Riemenschneider | Frederick Riemenschneider and Anette Frank | Exploring Large Language Models for Classical Philology | Paper accepted for publication at ACL 2023 Main; 10 pages, 7 appendix
pages, 4 figures, 13 tables | null | null | null | cs.CL | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Recent advances in NLP have led to the creation of powerful language models
for many languages including Ancient Greek and Latin. While prior work on
Classical languages unanimously uses BERT, in this work we create four language
models for Ancient Greek that vary along two dimensions to study their
versatility for tasks of interest for Classical languages: we explore (i)
encoder-only and encoder-decoder architectures using RoBERTa and T5 as strong
model types, and create for each of them (ii) a monolingual Ancient Greek and a
multilingual instance that includes Latin and English. We evaluate all models
on morphological and syntactic tasks, including lemmatization, which
demonstrates the added value of T5's decoding abilities. We further define two
probing tasks to investigate the knowledge acquired by models pre-trained on
Classical texts. Our experiments provide the first benchmarking analysis of
existing models of Ancient Greek. Results show that our models provide
significant improvements over the SoTA. The systematic analysis of model types
can inform future research in designing language models for Classical
languages, including the development of novel generative tasks. We make all our
models available as community resources, along with a large curated
pre-training corpus for Ancient Greek, to support the creation of a larger,
comparable model zoo for Classical Philology. Our models and resources are
available at https://github.com/Heidelberg-NLP/ancient-language-models.
| [
{
"version": "v1",
"created": "Tue, 23 May 2023 05:21:02 GMT"
}
] | 2023-05-24T00:00:00 | [
[
"Riemenschneider",
"Frederick",
""
],
[
"Frank",
"Anette",
""
]
] | TITLE: Exploring Large Language Models for Classical Philology
ABSTRACT: Recent advances in NLP have led to the creation of powerful language models
for many languages including Ancient Greek and Latin. While prior work on
Classical languages unanimously uses BERT, in this work we create four language
models for Ancient Greek that vary along two dimensions to study their
versatility for tasks of interest for Classical languages: we explore (i)
encoder-only and encoder-decoder architectures using RoBERTa and T5 as strong
model types, and create for each of them (ii) a monolingual Ancient Greek and a
multilingual instance that includes Latin and English. We evaluate all models
on morphological and syntactic tasks, including lemmatization, which
demonstrates the added value of T5's decoding abilities. We further define two
probing tasks to investigate the knowledge acquired by models pre-trained on
Classical texts. Our experiments provide the first benchmarking analysis of
existing models of Ancient Greek. Results show that our models provide
significant improvements over the SoTA. The systematic analysis of model types
can inform future research in designing language models for Classical
languages, including the development of novel generative tasks. We make all our
models available as community resources, along with a large curated
pre-training corpus for Ancient Greek, to support the creation of a larger,
comparable model zoo for Classical Philology. Our models and resources are
available at https://github.com/Heidelberg-NLP/ancient-language-models.
| no_new_dataset | 0.706564 |
2305.13700 | Chenglong Wang | Chenglong Wang, Jiangyan Yi, Jianhua Tao, Chuyuan Zhang, Shuai Zhang
and Xun Chen | Detection of Cross-Dataset Fake Audio Based on Prosodic and
Pronunciation Features | Interspeech2023 | null | null | null | cs.SD eess.AS | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Existing fake audio detection systems perform well in in-domain testing, but
still face many challenges in out-of-domain testing. This is due to the
mismatch between the training and test data, as well as the poor
generalizability of features extracted from limited views. To address this, we
propose multi-view features for fake audio detection, which aim to capture more
generalized features from prosodic, pronunciation, and wav2vec dimensions.
Specifically, the phoneme duration features are extracted from a pre-trained
model based on a large amount of speech data. For the pronunciation features, a
Conformer-based phoneme recognition model is first trained, keeping the
acoustic encoder part as a deeply embedded feature extractor. Furthermore, the
prosodic and pronunciation features are fused with wav2vec features based on an
attention mechanism to improve the generalization of fake audio detection
models. Results show that the proposed approach achieves significant
performance gains in several cross-dataset experiments.
| [
{
"version": "v1",
"created": "Tue, 23 May 2023 05:27:39 GMT"
}
] | 2023-05-24T00:00:00 | [
[
"Wang",
"Chenglong",
""
],
[
"Yi",
"Jiangyan",
""
],
[
"Tao",
"Jianhua",
""
],
[
"Zhang",
"Chuyuan",
""
],
[
"Zhang",
"Shuai",
""
],
[
"Chen",
"Xun",
""
]
] | TITLE: Detection of Cross-Dataset Fake Audio Based on Prosodic and
Pronunciation Features
ABSTRACT: Existing fake audio detection systems perform well in in-domain testing, but
still face many challenges in out-of-domain testing. This is due to the
mismatch between the training and test data, as well as the poor
generalizability of features extracted from limited views. To address this, we
propose multi-view features for fake audio detection, which aim to capture more
generalized features from prosodic, pronunciation, and wav2vec dimensions.
Specifically, the phoneme duration features are extracted from a pre-trained
model based on a large amount of speech data. For the pronunciation features, a
Conformer-based phoneme recognition model is first trained, keeping the
acoustic encoder part as a deeply embedded feature extractor. Furthermore, the
prosodic and pronunciation features are fused with wav2vec features based on an
attention mechanism to improve the generalization of fake audio detection
models. Results show that the proposed approach achieves significant
performance gains in several cross-dataset experiments.
| no_new_dataset | 0.707771 |
2305.13703 | Vered Shwartz | EunJeong Hwang and Vered Shwartz | MemeCap: A Dataset for Captioning and Interpreting Memes | null | null | null | null | cs.CL | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Memes are a widely popular tool for web users to express their thoughts using
visual metaphors. Understanding memes requires recognizing and interpreting
visual metaphors with respect to the text inside or around the meme, often
while employing background knowledge and reasoning abilities. We present the
task of meme captioning and release a new dataset, MemeCap. Our dataset
contains 6.3K memes along with the title of the post containing the meme, the
meme captions, the literal image caption, and the visual metaphors. Despite the
recent success of vision and language (VL) models on tasks such as image
captioning and visual question answering, our extensive experiments using
state-of-the-art VL models show that they still struggle with visual metaphors,
and perform substantially worse than humans.
| [
{
"version": "v1",
"created": "Tue, 23 May 2023 05:41:18 GMT"
}
] | 2023-05-24T00:00:00 | [
[
"Hwang",
"EunJeong",
""
],
[
"Shwartz",
"Vered",
""
]
] | TITLE: MemeCap: A Dataset for Captioning and Interpreting Memes
ABSTRACT: Memes are a widely popular tool for web users to express their thoughts using
visual metaphors. Understanding memes requires recognizing and interpreting
visual metaphors with respect to the text inside or around the meme, often
while employing background knowledge and reasoning abilities. We present the
task of meme captioning and release a new dataset, MemeCap. Our dataset
contains 6.3K memes along with the title of the post containing the meme, the
meme captions, the literal image caption, and the visual metaphors. Despite the
recent success of vision and language (VL) models on tasks such as image
captioning and visual question answering, our extensive experiments using
state-of-the-art VL models show that they still struggle with visual metaphors,
and perform substantially worse than humans.
| new_dataset | 0.71446 |
2305.13705 | Lijun Li | Lijun Li, Li'an Zhuo, Bang Zhang, Liefeng Bo, Chen Chen | DiffHand: End-to-End Hand Mesh Reconstruction via Diffusion Models | null | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Hand mesh reconstruction from the monocular image is a challenging task due
to its depth ambiguity and severe occlusion, there remains a non-unique mapping
between the monocular image and hand mesh. To address this, we develop
DiffHand, the first diffusion-based framework that approaches hand mesh
reconstruction as a denoising diffusion process. Our one-stage pipeline
utilizes noise to model the uncertainty distribution of the intermediate hand
mesh in a forward process. We reformulate the denoising diffusion process to
gradually refine noisy hand mesh and then select mesh with the highest
probability of being correct based on the image itself, rather than relying on
2D joints extracted beforehand. To better model the connectivity of hand
vertices, we design a novel network module called the cross-modality decoder.
Extensive experiments on the popular benchmarks demonstrate that our method
outperforms the state-of-the-art hand mesh reconstruction approaches by
achieving 5.8mm PA-MPJPE on the Freihand test set, 4.98mm PA-MPJPE on the
DexYCB test set.
| [
{
"version": "v1",
"created": "Tue, 23 May 2023 05:44:03 GMT"
}
] | 2023-05-24T00:00:00 | [
[
"Li",
"Lijun",
""
],
[
"Zhuo",
"Li'an",
""
],
[
"Zhang",
"Bang",
""
],
[
"Bo",
"Liefeng",
""
],
[
"Chen",
"Chen",
""
]
] | TITLE: DiffHand: End-to-End Hand Mesh Reconstruction via Diffusion Models
ABSTRACT: Hand mesh reconstruction from the monocular image is a challenging task due
to its depth ambiguity and severe occlusion, there remains a non-unique mapping
between the monocular image and hand mesh. To address this, we develop
DiffHand, the first diffusion-based framework that approaches hand mesh
reconstruction as a denoising diffusion process. Our one-stage pipeline
utilizes noise to model the uncertainty distribution of the intermediate hand
mesh in a forward process. We reformulate the denoising diffusion process to
gradually refine noisy hand mesh and then select mesh with the highest
probability of being correct based on the image itself, rather than relying on
2D joints extracted beforehand. To better model the connectivity of hand
vertices, we design a novel network module called the cross-modality decoder.
Extensive experiments on the popular benchmarks demonstrate that our method
outperforms the state-of-the-art hand mesh reconstruction approaches by
achieving 5.8mm PA-MPJPE on the Freihand test set, 4.98mm PA-MPJPE on the
DexYCB test set.
| no_new_dataset | 0.710019 |
2305.13707 | Orevaoghene Ahia | Orevaoghene Ahia, Sachin Kumar, Hila Gonen, Jungo Kasai, David R.
Mortensen, Noah A. Smith, Yulia Tsvetkov | Do All Languages Cost the Same? Tokenization in the Era of Commercial
Language Models | null | null | null | null | cs.CL | http://creativecommons.org/licenses/by/4.0/ | Language models have graduated from being research prototypes to
commercialized products offered as web APIs, and recent works have highlighted
the multilingual capabilities of these products. The API vendors charge their
users based on usage, more specifically on the number of ``tokens'' processed
or generated by the underlying language models. What constitutes a token,
however, is training data and model dependent with a large variance in the
number of tokens required to convey the same information in different
languages. In this work, we analyze the effect of this non-uniformity on the
fairness of an API's pricing policy across languages. We conduct a systematic
analysis of the cost and utility of OpenAI's language model API on multilingual
benchmarks in 22 typologically diverse languages. We show evidence that
speakers of a large number of the supported languages are overcharged while
obtaining poorer results. These speakers tend to also come from regions where
the APIs are less affordable to begin with. Through these analyses, we aim to
increase transparency around language model APIs' pricing policies and
encourage the vendors to make them more equitable.
| [
{
"version": "v1",
"created": "Tue, 23 May 2023 05:46:45 GMT"
}
] | 2023-05-24T00:00:00 | [
[
"Ahia",
"Orevaoghene",
""
],
[
"Kumar",
"Sachin",
""
],
[
"Gonen",
"Hila",
""
],
[
"Kasai",
"Jungo",
""
],
[
"Mortensen",
"David R.",
""
],
[
"Smith",
"Noah A.",
""
],
[
"Tsvetkov",
"Yulia",
""
]
] | TITLE: Do All Languages Cost the Same? Tokenization in the Era of Commercial
Language Models
ABSTRACT: Language models have graduated from being research prototypes to
commercialized products offered as web APIs, and recent works have highlighted
the multilingual capabilities of these products. The API vendors charge their
users based on usage, more specifically on the number of ``tokens'' processed
or generated by the underlying language models. What constitutes a token,
however, is training data and model dependent with a large variance in the
number of tokens required to convey the same information in different
languages. In this work, we analyze the effect of this non-uniformity on the
fairness of an API's pricing policy across languages. We conduct a systematic
analysis of the cost and utility of OpenAI's language model API on multilingual
benchmarks in 22 typologically diverse languages. We show evidence that
speakers of a large number of the supported languages are overcharged while
obtaining poorer results. These speakers tend to also come from regions where
the APIs are less affordable to begin with. Through these analyses, we aim to
increase transparency around language model APIs' pricing policies and
encourage the vendors to make them more equitable.
| no_new_dataset | 0.712045 |
2305.13710 | Qingyang Wu | Qingyang Wu, Deema Alnuhait, Derek Chen, Zhou Yu | Using Textual Interface to Align External Knowledge for End-to-End
Task-Oriented Dialogue Systems | null | null | null | null | cs.CL | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Traditional end-to-end task-oriented dialogue systems have been built with a
modularized design. However, such design often causes misalignment between the
agent response and external knowledge, due to inadequate representation of
information. Furthermore, its evaluation metrics emphasize assessing the
agent's pre-lexicalization response, neglecting the quality of the completed
response. In this work, we propose a novel paradigm that uses a textual
interface to align external knowledge and eliminate redundant processes. We
demonstrate our paradigm in practice through MultiWOZ-Remake, including an
interactive textual interface built for the MultiWOZ database and a
correspondingly re-processed dataset. We train an end-to-end dialogue system to
evaluate this new dataset. The experimental results show that our approach
generates more natural final responses and achieves a greater task success rate
compared to the previous models.
| [
{
"version": "v1",
"created": "Tue, 23 May 2023 05:48:21 GMT"
}
] | 2023-05-24T00:00:00 | [
[
"Wu",
"Qingyang",
""
],
[
"Alnuhait",
"Deema",
""
],
[
"Chen",
"Derek",
""
],
[
"Yu",
"Zhou",
""
]
] | TITLE: Using Textual Interface to Align External Knowledge for End-to-End
Task-Oriented Dialogue Systems
ABSTRACT: Traditional end-to-end task-oriented dialogue systems have been built with a
modularized design. However, such design often causes misalignment between the
agent response and external knowledge, due to inadequate representation of
information. Furthermore, its evaluation metrics emphasize assessing the
agent's pre-lexicalization response, neglecting the quality of the completed
response. In this work, we propose a novel paradigm that uses a textual
interface to align external knowledge and eliminate redundant processes. We
demonstrate our paradigm in practice through MultiWOZ-Remake, including an
interactive textual interface built for the MultiWOZ database and a
correspondingly re-processed dataset. We train an end-to-end dialogue system to
evaluate this new dataset. The experimental results show that our approach
generates more natural final responses and achieves a greater task success rate
compared to the previous models.
| no_new_dataset | 0.698471 |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.