paper_url
stringlengths 36
81
| paper_title
stringlengths 1
242
⌀ | paper_arxiv_id
stringlengths 9
16
⌀ | paper_url_abs
stringlengths 18
314
| paper_url_pdf
stringlengths 21
935
⌀ | repo_url
stringlengths 26
200
| is_official
bool 2
classes | mentioned_in_paper
bool 2
classes | mentioned_in_github
bool 2
classes | framework
stringclasses 9
values |
---|---|---|---|---|---|---|---|---|---|
https://paperswithcode.com/paper/fedentropy-efficient-device-grouping-for
|
FedEntropy: Efficient Device Grouping for Federated Learning Using Maximum Entropy Judgment
|
2205.12038
|
https://arxiv.org/abs/2205.12038v1
|
https://arxiv.org/pdf/2205.12038v1.pdf
|
https://github.com/fedentropy/fedentropy
| true | true | false |
pytorch
|
https://paperswithcode.com/paper/learning-transferable-visual-models-from
|
Learning Transferable Visual Models From Natural Language Supervision
|
2103.00020
|
https://arxiv.org/abs/2103.00020v1
|
https://arxiv.org/pdf/2103.00020v1.pdf
|
https://github.com/sberbank-ai/ru-clip
| false | false | true |
pytorch
|
https://paperswithcode.com/paper/tackling-fake-news-detection-by-continually
|
Tackling Fake News Detection by Continually Improving Social Context Representations using Graph Neural Networks
| null |
https://aclanthology.org/2022.acl-long.97
|
https://aclanthology.org/2022.acl-long.97.pdf
|
https://github.com/hockeybro12/fakenews_inference_operators
| true | true | false |
pytorch
|
https://paperswithcode.com/paper/bayesian-functional-principal-components-1
|
Bayesian modeling of nearly mutually orthogonal processes
|
2205.12361
|
https://arxiv.org/abs/2205.12361v3
|
https://arxiv.org/pdf/2205.12361v3.pdf
|
https://github.com/jamesmatuk/remo-fpca
| true | true | false |
none
|
https://paperswithcode.com/paper/bionic-tracking-using-eye-tracking-to-track
|
Bionic Tracking: Using Eye Tracking to Track Biological Cells in Virtual Reality
|
2005.00387
|
https://arxiv.org/abs/2005.00387v2
|
https://arxiv.org/pdf/2005.00387v2.pdf
|
https://github.com/scenerygraphics/bionic-tracking
| true | true | false |
none
|
https://paperswithcode.com/paper/the-neuro-symbolic-brain
|
The Neuro-Symbolic Brain
|
2205.13440
|
https://arxiv.org/abs/2205.13440v1
|
https://arxiv.org/pdf/2205.13440v1.pdf
|
https://github.com/robertlizee/neuro-symbolic-vm
| true | false | true |
none
|
https://paperswithcode.com/paper/flexible-and-fast-estimation-of-binary-merger
|
Flexible and Fast Estimation of Binary Merger Population Distributions with Adaptive KDE
|
2112.12659
|
https://arxiv.org/abs/2112.12659v3
|
https://arxiv.org/pdf/2112.12659v3.pdf
|
https://github.com/jamsadiq/peakdetectionalgorithm
| true | true | true |
none
|
https://paperswithcode.com/paper/domain-adaptive-faster-r-cnn-for-object
|
Domain Adaptive Faster R-CNN for Object Detection in the Wild
|
1803.03243
|
http://arxiv.org/abs/1803.03243v1
|
http://arxiv.org/pdf/1803.03243v1.pdf
|
https://github.com/shreyasrajesh/DA-Object-Detection
| false | false | true |
pytorch
|
https://paperswithcode.com/paper/cross-modality-discrepant-interaction-network
|
Cross-modality Discrepant Interaction Network for RGB-D Salient Object Detection
|
2108.01971
|
https://arxiv.org/abs/2108.01971v1
|
https://arxiv.org/pdf/2108.01971v1.pdf
|
https://github.com/kingcong/CDINet
| false | false | false |
mindspore
|
https://paperswithcode.com/paper/extending-the-design-space-of-graph-neural-1
|
Extending the Design Space of Graph Neural Networks by Rethinking Folklore Weisfeiler-Lehman
|
2306.03266
|
https://arxiv.org/abs/2306.03266v3
|
https://arxiv.org/pdf/2306.03266v3.pdf
|
https://github.com/jiaruifeng/n2gnn
| true | true | true |
pytorch
|
https://paperswithcode.com/paper/transfuser-imitation-with-transformer-based
|
TransFuser: Imitation with Transformer-Based Sensor Fusion for Autonomous Driving
|
2205.15997
|
https://arxiv.org/abs/2205.15997v1
|
https://arxiv.org/pdf/2205.15997v1.pdf
|
https://github.com/autonomousvision/transfuser
| true | true | true |
pytorch
|
https://paperswithcode.com/paper/a-unified-weight-initialization-paradigm-for
|
A Unified Weight Initialization Paradigm for Tensorial Convolutional Neural Networks
|
2205.15307
|
https://arxiv.org/abs/2205.15307v2
|
https://arxiv.org/pdf/2205.15307v2.pdf
|
https://github.com/tnbar/tednet
| true | true | false |
pytorch
|
https://paperswithcode.com/paper/preparing-an-endangered-language-for-the
|
Preparing an Endangered Language for the Digital Age: The Case of Judeo-Spanish
|
2205.15599
|
https://arxiv.org/abs/2205.15599v1
|
https://arxiv.org/pdf/2205.15599v1.pdf
|
https://github.com/collectivat-dev/espanyol-ladino-translation
| true | true | false |
none
|
https://paperswithcode.com/paper/aggregated-residual-transformations-for-deep
|
Aggregated Residual Transformations for Deep Neural Networks
|
1611.05431
|
http://arxiv.org/abs/1611.05431v2
|
http://arxiv.org/pdf/1611.05431v2.pdf
|
https://github.com/2023-MindSpore-1/ms-code-13
| false | false | false |
mindspore
|
https://paperswithcode.com/paper/a-theoretical-study-on-solving-continual
|
A Theoretical Study on Solving Continual Learning
|
2211.02633
|
https://arxiv.org/abs/2211.02633v1
|
https://arxiv.org/pdf/2211.02633v1.pdf
|
https://github.com/k-gyuhak/wptp
| true | true | false |
pytorch
|
https://paperswithcode.com/paper/an-image-is-worth-16x16-words-transformers-1
|
An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale
|
2010.11929
|
https://arxiv.org/abs/2010.11929v2
|
https://arxiv.org/pdf/2010.11929v2.pdf
|
https://github.com/OML-Team/open-metric-learning
| false | false | false |
pytorch
|
https://paperswithcode.com/paper/constrained-variational-policy-optimization
|
Constrained Variational Policy Optimization for Safe Reinforcement Learning
|
2201.11927
|
https://arxiv.org/abs/2201.11927v3
|
https://arxiv.org/pdf/2201.11927v3.pdf
|
https://github.com/liuzuxin/cvpo-safe-rl
| true | true | true |
pytorch
|
https://paperswithcode.com/paper/pgmpy-a-python-toolkit-for-bayesian-networks
|
pgmpy: A Python Toolkit for Bayesian Networks
|
2304.08639
|
https://arxiv.org/abs/2304.08639v1
|
https://arxiv.org/pdf/2304.08639v1.pdf
|
https://github.com/pgmpy/pgmpy
| true | true | false |
pytorch
|
https://paperswithcode.com/paper/cyclemix-a-holistic-strategy-for-medical
|
CycleMix: A Holistic Strategy for Medical Image Segmentation from Scribble Supervision
|
2203.01475
|
https://arxiv.org/abs/2203.01475v2
|
https://arxiv.org/pdf/2203.01475v2.pdf
|
https://github.com/bwgzk/cyclemix
| true | true | true |
pytorch
|
https://paperswithcode.com/paper/recbole-2-0-towards-a-more-up-to-date
|
RecBole 2.0: Towards a More Up-to-Date Recommendation Library
|
2206.07351
|
https://arxiv.org/abs/2206.07351v2
|
https://arxiv.org/pdf/2206.07351v2.pdf
|
https://github.com/rucaibox/recbole2.0
| true | true | true |
pytorch
|
https://paperswithcode.com/paper/emerging-properties-in-self-supervised-vision
|
Emerging Properties in Self-Supervised Vision Transformers
|
2104.14294
|
https://arxiv.org/abs/2104.14294v2
|
https://arxiv.org/pdf/2104.14294v2.pdf
|
https://github.com/OML-Team/open-metric-learning
| false | false | false |
pytorch
|
https://paperswithcode.com/paper/bevdet-high-performance-multi-camera-3d
|
BEVDet: High-performance Multi-camera 3D Object Detection in Bird-Eye-View
|
2112.11790
|
https://arxiv.org/abs/2112.11790v3
|
https://arxiv.org/pdf/2112.11790v3.pdf
|
https://github.com/HuangJunJie2017/BEVDet
| true | true | true |
pytorch
|
https://paperswithcode.com/paper/on-the-surprising-behaviour-of-node2vec
|
On the Surprising Behaviour of node2vec
|
2206.08252
|
https://arxiv.org/abs/2206.08252v2
|
https://arxiv.org/pdf/2206.08252v2.pdf
|
https://github.com/aidos-lab/node2vec-surprises
| true | true | false |
pytorch
|
https://paperswithcode.com/paper/distributionally-robust-losses-for-latent
|
Distributionally Robust Losses for Latent Covariate Mixtures
|
2007.13982
|
https://arxiv.org/abs/2007.13982v2
|
https://arxiv.org/pdf/2007.13982v2.pdf
|
https://github.com/hsnamkoong/marginal-dro
| true | true | false |
pytorch
|
https://paperswithcode.com/paper/projection-scrubbing-a-more-effective-data
|
Less is more: balancing noise reduction and data retention in fMRI with data-driven scrubbing
|
2108.00319
|
https://arxiv.org/abs/2108.00319v4
|
https://arxiv.org/pdf/2108.00319v4.pdf
|
https://github.com/cran/fMRIscrub
| false | false | true |
none
|
https://paperswithcode.com/paper/learning-stochastic-parametric-differentiable
|
Learning Stochastic Parametric Differentiable Predictive Control Policies
|
2203.01447
|
https://arxiv.org/abs/2203.01447v2
|
https://arxiv.org/pdf/2203.01447v2.pdf
|
https://github.com/pnnl/neuromancer
| true | true | false |
pytorch
|
https://paperswithcode.com/paper/neural-inverse-kinematics
|
Neural Inverse Kinematics
|
2205.10837
|
https://arxiv.org/abs/2205.10837v1
|
https://arxiv.org/pdf/2205.10837v1.pdf
|
https://github.com/RaphaelBensTAU/NeuralInverseKinematics
| true | false | true |
pytorch
|
https://paperswithcode.com/paper/instant-neural-graphics-primitives-with-a
|
Instant Neural Graphics Primitives with a Multiresolution Hash Encoding
|
2201.05989
|
https://arxiv.org/abs/2201.05989v2
|
https://arxiv.org/pdf/2201.05989v2.pdf
|
https://github.com/kair-bair/nerfacc
| false | false | false |
pytorch
|
https://paperswithcode.com/paper/the-sigma-8-tension-is-a-drag
|
The Sigma-8 Tension is a Drag
|
2209.06217
|
https://arxiv.org/abs/2209.06217v2
|
https://arxiv.org/pdf/2209.06217v2.pdf
|
https://github.com/brinckmann/montepython_public
| true | true | false |
none
|
https://paperswithcode.com/paper/automatic-correction-of-human-translations
|
Automatic Correction of Human Translations
|
2206.08593
|
https://arxiv.org/abs/2206.08593v1
|
https://arxiv.org/pdf/2206.08593v1.pdf
|
https://github.com/lilt/tec
| true | true | true |
none
|
https://paperswithcode.com/paper/thompson-sampling-for-robust-transfer-in
|
Thompson Sampling for Robust Transfer in Multi-Task Bandits
|
2206.08556
|
https://arxiv.org/abs/2206.08556v1
|
https://arxiv.org/pdf/2206.08556v1.pdf
|
https://github.com/zhiwang123/eps-mpmab-ts
| true | true | false |
none
|
https://paperswithcode.com/paper/object-structural-points-representation-for
|
Object Structural Points Representation for Graph-based Semantic Monocular Localization and Mapping
|
2206.10263
|
https://arxiv.org/abs/2206.10263v1
|
https://arxiv.org/pdf/2206.10263v1.pdf
|
https://github.com/airlab-polimi/c-slam
| true | true | false |
none
|
https://paperswithcode.com/paper/diagnostic-tool-for-out-of-sample-model
|
Diagnostic Tool for Out-of-Sample Model Evaluation
|
2206.10982
|
https://arxiv.org/abs/2206.10982v3
|
https://arxiv.org/pdf/2206.10982v3.pdf
|
https://github.com/el-hult/lal
| true | true | false |
pytorch
|
https://paperswithcode.com/paper/panoramic-panoptic-segmentation-insights-into
|
Panoramic Panoptic Segmentation: Insights Into Surrounding Parsing for Mobile Agents via Unsupervised Contrastive Learning
|
2206.10711
|
https://arxiv.org/abs/2206.10711v2
|
https://arxiv.org/pdf/2206.10711v2.pdf
|
https://github.com/alexanderjaus/PPS
| true | true | true |
pytorch
|
https://paperswithcode.com/paper/a-novel-approach-for-exploring-the-light
|
A Novel Approach for Exploring the Light Traveling Path in the Medium with a Spherically Symmetric Refractive Index
|
2212.02642
|
https://arxiv.org/abs/2212.02642v1
|
https://arxiv.org/pdf/2212.02642v1.pdf
|
https://github.com/shengyangzhuang/A-Novel-Approach-for-Exploring-the-Light-Traveling-Path-in-the-Spherically-Symmetric-Medium
| true | false | false |
none
|
https://paperswithcode.com/paper/symmetric-network-with-spatial-relationship
|
Symmetric Network with Spatial Relationship Modeling for Natural Language-based Vehicle Retrieval
|
2206.10879
|
https://arxiv.org/abs/2206.10879v1
|
https://arxiv.org/pdf/2206.10879v1.pdf
|
https://github.com/hbchen121/aicity2022_track2_ssm
| true | true | true |
pytorch
|
https://paperswithcode.com/paper/visfis-visual-feature-importance-supervision
|
VisFIS: Visual Feature Importance Supervision with Right-for-the-Right-Reason Objectives
|
2206.11212
|
https://arxiv.org/abs/2206.11212v2
|
https://arxiv.org/pdf/2206.11212v2.pdf
|
https://github.com/zfying/visfis
| true | true | true |
pytorch
|
https://paperswithcode.com/paper/matrix-completion-and-low-rank-svd-via-fast
|
Matrix Completion and Low-Rank SVD via Fast Alternating Least Squares
|
1410.2596
|
http://arxiv.org/abs/1410.2596v1
|
http://arxiv.org/pdf/1410.2596v1.pdf
|
https://github.com/travisbrady/py-soft-impute
| false | false | true |
none
|
https://paperswithcode.com/paper/block-diffusion-interpolating-between
|
Block Diffusion: Interpolating Between Autoregressive and Diffusion Language Models
|
2503.09573
|
https://arxiv.org/abs/2503.09573v3
|
https://arxiv.org/pdf/2503.09573v3.pdf
|
https://github.com/MindSpore-scientific/code-12/tree/main/Block_Model
| false | false | false |
mindspore
|
https://paperswithcode.com/paper/a-simple-and-efficient-sampling-based
|
A Simple and Efficient Sampling-based Algorithm for General Reachability Analysis
|
2112.05745
|
https://arxiv.org/abs/2112.05745v3
|
https://arxiv.org/pdf/2112.05745v3.pdf
|
https://github.com/stanfordasl/stochasticedl
| false | false | true |
jax
|
https://paperswithcode.com/paper/a-multi-head-model-for-continual-learning-via
|
A Multi-Head Model for Continual Learning via Out-of-Distribution Replay
|
2208.09734
|
https://arxiv.org/abs/2208.09734v1
|
https://arxiv.org/pdf/2208.09734v1.pdf
|
https://github.com/k-gyuhak/wptp
| false | false | true |
pytorch
|
https://paperswithcode.com/paper/u-net-convolutional-networks-for-biomedical
|
U-Net: Convolutional Networks for Biomedical Image Segmentation
|
1505.04597
|
http://arxiv.org/abs/1505.04597v1
|
http://arxiv.org/pdf/1505.04597v1.pdf
|
https://github.com/udacity/MLND-CN-Capstone-TGSImage
| false | false | true |
none
|
https://paperswithcode.com/paper/rethinking-atrous-convolution-for-semantic
|
Rethinking Atrous Convolution for Semantic Image Segmentation
|
1706.05587
|
http://arxiv.org/abs/1706.05587v3
|
http://arxiv.org/pdf/1706.05587v3.pdf
|
https://github.com/udacity/MLND-CN-Capstone-TGSImage
| false | false | true |
none
|
https://paperswithcode.com/paper/image-aesthetics-assessment-using-graph
|
Image Aesthetics Assessment Using Graph Attention Network
|
2206.12869
|
https://arxiv.org/abs/2206.12869v2
|
https://arxiv.org/pdf/2206.12869v2.pdf
|
https://github.com/koustav123/aesthetics_assessment_using_graphs
| true | true | false |
pytorch
|
https://paperswithcode.com/paper/rethinking-cnn-models-for-audio
|
Rethinking CNN Models for Audio Classification
|
2007.11154
|
https://arxiv.org/abs/2007.11154v2
|
https://arxiv.org/pdf/2007.11154v2.pdf
|
https://github.com/shijing001/unicertainty_calibration_audio_classifiers
| false | false | true |
pytorch
|
https://paperswithcode.com/paper/uncertainty-calibration-for-deep-audio
|
Uncertainty Calibration for Deep Audio Classifiers
|
2206.13071
|
https://arxiv.org/abs/2206.13071v1
|
https://arxiv.org/pdf/2206.13071v1.pdf
|
https://github.com/shijing001/unicertainty_calibration_audio_classifiers
| true | true | true |
pytorch
|
https://paperswithcode.com/paper/enhancing-stochastic-petri-net-based
|
Enhancing Stochastic Petri Net-based Remaining Time Prediction using k-Nearest Neighbors
|
2206.13109
|
https://arxiv.org/abs/2206.13109v1
|
https://arxiv.org/pdf/2206.13109v1.pdf
|
https://github.com/jarnevdb/bp-time-prediction-using-knn
| true | true | false |
none
|
https://paperswithcode.com/paper/190807919
|
Deep High-Resolution Representation Learning for Visual Recognition
|
1908.07919
|
https://arxiv.org/abs/1908.07919v2
|
https://arxiv.org/pdf/1908.07919v2.pdf
|
https://github.com/kingcong/gpu_HRNetW48_cls
| false | false | true |
mindspore
|
https://paperswithcode.com/paper/feature-overcorrelation-in-deep-graph-neural
|
Feature Overcorrelation in Deep Graph Neural Networks: A New Perspective
|
2206.07743
|
https://arxiv.org/abs/2206.07743v1
|
https://arxiv.org/pdf/2206.07743v1.pdf
|
https://github.com/chandlerbang/decorr
| true | true | true |
pytorch
|
https://paperswithcode.com/paper/towards-overcoming-data-scarcity-in-materials
|
Towards overcoming data scarcity in materials science: unifying models and datasets with a mixture of experts framework
|
2207.13880
|
https://arxiv.org/abs/2207.13880v1
|
https://arxiv.org/pdf/2207.13880v1.pdf
|
https://github.com/rees-c/moe
| true | true | false |
pytorch
|
https://paperswithcode.com/paper/direct-preference-optimization-your-language
|
Direct Preference Optimization: Your Language Model is Secretly a Reward Model
|
2305.18290
|
https://arxiv.org/abs/2305.18290v3
|
https://arxiv.org/pdf/2305.18290v3.pdf
|
https://github.com/KomeijiForce/Active_Passive_Constraint_Koishiday_2024
| false | false | true |
pytorch
|
https://paperswithcode.com/paper/ego-planner-an-esdf-free-gradient-based-local
|
EGO-Planner: An ESDF-free Gradient-based Local Planner for Quadrotors
|
2008.08835
|
https://arxiv.org/abs/2008.08835v1
|
https://arxiv.org/pdf/2008.08835v1.pdf
|
https://github.com/j-marple-dev/ego-planner
| false | false | true |
none
|
https://paperswithcode.com/paper/learning-maritime-obstacle-detection-from
|
Learning Maritime Obstacle Detection from Weak Annotations by Scaffolding
|
2108.00564
|
https://arxiv.org/abs/2108.00564v1
|
https://arxiv.org/pdf/2108.00564v1.pdf
|
https://github.com/lojzezust/slr
| false | false | true |
pytorch
|
https://paperswithcode.com/paper/sparse-distillation-speeding-up-text
|
Sparse Distillation: Speeding Up Text Classification by Using Bigger Student Models
|
2110.08536
|
https://arxiv.org/abs/2110.08536v2
|
https://arxiv.org/pdf/2110.08536v2.pdf
|
https://github.com/ink-usc/sparse-distillation
| true | true | true |
pytorch
|
https://paperswithcode.com/paper/in-defense-of-the-triplet-loss-for-person-re
|
In Defense of the Triplet Loss for Person Re-Identification
|
1703.07737
|
http://arxiv.org/abs/1703.07737v4
|
http://arxiv.org/pdf/1703.07737v4.pdf
|
https://github.com/OML-Team/open-metric-learning
| false | false | false |
pytorch
|
https://paperswithcode.com/paper/non-gaussianity-in-cmb-lensing-from-full-sky
|
Non-Gaussianity in CMB lensing from full-sky simulations
|
2411.02774
|
https://arxiv.org/abs/2411.02774v3
|
https://arxiv.org/pdf/2411.02774v3.pdf
|
https://github.com/Kang-Yuqi/FLAReS
| true | false | false |
none
|
https://paperswithcode.com/paper/serendipity-in-dark-photon-searches
|
Serendipity in dark photon searches
|
1801.04847
|
https://arxiv.org/abs/1801.04847v2
|
https://arxiv.org/pdf/1801.04847v2.pdf
|
https://gitlab.com/philten/darkcast
| true | true | true |
none
|
https://paperswithcode.com/paper/transformcode-a-contrastive-learning
|
TransformCode: A Contrastive Learning Framework for Code Embedding via Subtree Transformation
|
2311.08157
|
https://arxiv.org/abs/2311.08157v2
|
https://arxiv.org/pdf/2311.08157v2.pdf
|
https://github.com/iamfaith/transformcode
| true | true | false |
pytorch
|
https://paperswithcode.com/paper/technical-report-large-language-models-can
|
Large Language Models can Strategically Deceive their Users when Put Under Pressure
|
2311.07590
|
https://arxiv.org/abs/2311.07590v4
|
https://arxiv.org/pdf/2311.07590v4.pdf
|
https://github.com/apolloresearch/insider-trading
| true | true | false |
none
|
https://paperswithcode.com/paper/detecting-covariate-drift-in-text-data-using
|
Detecting covariate drift in text data using document embeddings and dimensionality reduction
|
2309.10000
|
https://arxiv.org/abs/2309.10000v1
|
https://arxiv.org/pdf/2309.10000v1.pdf
|
https://github.com/vinayaksodar/nlp_drift_paper_code
| true | true | false |
none
|
https://paperswithcode.com/paper/adaptive-measurement-strategy-for-quantum
|
Adaptive measurement strategy for quantum subspace methods
|
2311.07893
|
https://arxiv.org/abs/2311.07893v2
|
https://arxiv.org/pdf/2311.07893v2.pdf
|
https://github.com/quantum-programming/adaptive-subspace
| true | false | false |
none
|
https://paperswithcode.com/paper/coatnet-marrying-convolution-and-attention
|
CoAtNet: Marrying Convolution and Attention for All Data Sizes
|
2106.04803
|
https://arxiv.org/abs/2106.04803v2
|
https://arxiv.org/pdf/2106.04803v2.pdf
|
https://github.com/MS-Mind/MS-Code-02/tree/main/configs/coat
| false | false | false |
mindspore
|
https://paperswithcode.com/paper/a-study-of-slang-representation-methods
|
A Study of Slang Representation Methods
|
2212.05613
|
https://arxiv.org/abs/2212.05613v3
|
https://arxiv.org/pdf/2212.05613v3.pdf
|
https://github.com/usc-isi-i2/slang-representation-learning
| true | false | false |
pytorch
|
https://paperswithcode.com/paper/traffic-state-data-imputation-an-efficient
|
Traffic state data imputation: An efficient generating method based on the graph aggregator
| null |
https://ieeexplore.ieee.org/abstract/document/9582618
|
https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=9582618
|
https://github.com/pihang/GA-GAN
| true | false | false |
none
|
https://paperswithcode.com/paper/ask2transformers-zero-shot-domain-labelling-1
|
Ask2Transformers: Zero-Shot Domain labelling with Pretrained Language Models
| null |
https://aclanthology.org/2021.gwc-1.6
|
https://aclanthology.org/2021.gwc-1.6.pdf
|
https://github.com/osainz59/Ask2Transformers
| true | true | false |
pytorch
|
https://paperswithcode.com/paper/a-field-test-of-bandit-algorithms-for
|
A Field Test of Bandit Algorithms for Recommendations: Understanding the Validity of Assumptions on Human Preferences in Multi-armed Bandits
|
2304.09088
|
https://arxiv.org/abs/2304.09088v1
|
https://arxiv.org/pdf/2304.09088v1.pdf
|
https://github.com/humainlab/human-bandit-evaluation
| true | true | false |
none
|
https://paperswithcode.com/paper/probabilistic-embeddings-for-cross-modal
|
Probabilistic Embeddings for Cross-Modal Retrieval
|
2101.05068
|
https://arxiv.org/abs/2101.05068v2
|
https://arxiv.org/pdf/2101.05068v2.pdf
|
https://github.com/naver-ai/pcme
| true | true | true |
pytorch
|
https://paperswithcode.com/paper/computing-and-exploiting-document-structure
|
Computing and Exploiting Document Structure to Improve Unsupervised Extractive Summarization of Legal Case Decisions
|
2211.03229
|
https://arxiv.org/abs/2211.03229v1
|
https://arxiv.org/pdf/2211.03229v1.pdf
|
https://github.com/cs329yangzhong/documentstructurelegalsum
| true | true | false |
none
|
https://paperswithcode.com/paper/meta-networks
|
Meta Networks
|
1703.00837
|
http://arxiv.org/abs/1703.00837v2
|
http://arxiv.org/pdf/1703.00837v2.pdf
|
https://bitbucket.org/tsendeemts/metanet
| true | true | true |
none
|
https://paperswithcode.com/paper/model-predictive-control-of-nonlinear-latent
|
Model Predictive Control of Nonlinear Latent Force Models: A Scenario-Based Approach
|
2207.13872
|
https://arxiv.org/abs/2207.13872v1
|
https://arxiv.org/pdf/2207.13872v1.pdf
|
https://github.com/KU-ISSL/MPC-NLFM-Scenario-ICRA21
| true | false | false |
none
|
https://paperswithcode.com/paper/eccv-caption-correcting-false-negatives-by
|
ECCV Caption: Correcting False Negatives by Collecting Machine-and-Human-verified Image-Caption Associations for MS-COCO
|
2204.03359
|
https://arxiv.org/abs/2204.03359v5
|
https://arxiv.org/pdf/2204.03359v5.pdf
|
https://github.com/naver-ai/pcme
| false | false | true |
pytorch
|
https://paperswithcode.com/paper/4k-haze-a-dehazing-benchmark-with-4k
|
4K-HAZE: A Dehazing Benchmark with 4K Resolution Hazy and Haze-Free Images
|
2303.15848
|
https://arxiv.org/abs/2303.15848v1
|
https://arxiv.org/pdf/2303.15848v1.pdf
|
https://github.com/zzr-idam/4KDehazing
| true | true | false |
pytorch
|
https://paperswithcode.com/paper/foundationpose-unified-6d-pose-estimation-and
|
FoundationPose: Unified 6D Pose Estimation and Tracking of Novel Objects
|
2312.08344
|
https://arxiv.org/abs/2312.08344v2
|
https://arxiv.org/pdf/2312.08344v2.pdf
|
https://github.com/NVlabs/FoundationStereo
| true | false | false |
pytorch
|
https://paperswithcode.com/paper/a-generative-approach-for-script-event-1
|
A Generative Approach for Script Event Prediction via Contrastive Fine-tuning
|
2212.03496
|
https://arxiv.org/abs/2212.03496v3
|
https://arxiv.org/pdf/2212.03496v3.pdf
|
https://github.com/zhufq00/mcnc
| true | true | false |
pytorch
|
https://paperswithcode.com/paper/bootstrap-state-representation-using-style
|
Bootstrap State Representation using Style Transfer for Better Generalization in Deep Reinforcement Learning
|
2207.07749
|
https://arxiv.org/abs/2207.07749v1
|
https://arxiv.org/pdf/2207.07749v1.pdf
|
https://github.com/masud99r/thinker
| true | true | true |
pytorch
|
https://paperswithcode.com/paper/dual-branch-hybrid-learning-network-for
|
Dual-branch Hybrid Learning Network for Unbiased Scene Graph Generation
|
2207.07913
|
https://arxiv.org/abs/2207.07913v1
|
https://arxiv.org/pdf/2207.07913v1.pdf
|
https://github.com/aa200647963/sgg-dhl
| true | true | true |
pytorch
|
https://paperswithcode.com/paper/facial-expression-and-attributes-recognition-1
|
Facial expression and attributes recognition based on multi-task learning of lightweight neural networks
|
2103.17107
|
https://arxiv.org/abs/2103.17107v3
|
https://arxiv.org/pdf/2103.17107v3.pdf
|
https://github.com/tomas-gajarsky/facetorch
| false | false | false |
pytorch
|
https://paperswithcode.com/paper/slowly-varying-regression-under-sparsity
|
Slowly Varying Regression under Sparsity
|
2102.10773
|
https://arxiv.org/abs/2102.10773v5
|
https://arxiv.org/pdf/2102.10773v5.pdf
|
https://github.com/vvdigalakis/ssvregression
| true | true | false |
none
|
https://paperswithcode.com/paper/one-person-one-model-learning-compound-router
|
One Person, One Model--Learning Compound Router for Sequential Recommendation
|
2211.02824
|
https://arxiv.org/abs/2211.02824v2
|
https://arxiv.org/pdf/2211.02824v2.pdf
|
https://github.com/icantnamemyself/canet
| true | true | false |
pytorch
|
https://paperswithcode.com/paper/networked-federated-multi-task-learning
|
Clustered Federated Learning via Generalized Total Variation Minimization
|
2105.12769
|
https://arxiv.org/abs/2105.12769v4
|
https://arxiv.org/pdf/2105.12769v4.pdf
|
https://github.com/sahelyiyi/FederatedLearning
| true | true | true |
pytorch
|
https://paperswithcode.com/paper/improving-federated-learning-personalization-1
|
Improving Federated Learning Personalization via Model Agnostic Meta Learning
|
1909.12488
|
https://arxiv.org/abs/1909.12488v2
|
https://arxiv.org/pdf/1909.12488v2.pdf
|
https://github.com/xiuyu0000/new_papers_codes/tree/main/FSMAFL
| false | false | false |
mindspore
|
https://paperswithcode.com/paper/preconditioned-nonlinear-conjugate-gradient-2
|
Preconditioned Nonlinear Conjugate Gradient Method for Real-time Interior-point Hyperelasticity
|
2405.08001
|
https://arxiv.org/abs/2405.08001v1
|
https://arxiv.org/pdf/2405.08001v1.pdf
|
https://github.com/Xingbaji/PNCG_IPC
| true | false | false |
none
|
https://paperswithcode.com/paper/mfan-multi-modal-feature-enhanced-attention
|
MFAN: Multi-modal Feature-enhanced Attention Networks for Rumor Detection
| null |
https://www.ijcai.org/proceedings/2022/335
|
https://www.ijcai.org/proceedings/2022/0335.pdf
|
https://github.com/drivsaf/MFAN
| false | false | false |
pytorch
|
https://paperswithcode.com/paper/bit-depth-enhancement-detection-for
|
Bit-depth enhancement detection for compressed video
|
2211.04799
|
https://arxiv.org/abs/2211.04799v1
|
https://arxiv.org/pdf/2211.04799v1.pdf
|
https://github.com/msu-video-group/bdedm
| true | true | false |
none
|
https://paperswithcode.com/paper/a-hierarchical-semantic-segmentation
|
A hierarchical semantic segmentation framework for computer vision-based bridge damage detection
|
2207.08878
|
https://arxiv.org/abs/2207.08878v2
|
https://arxiv.org/pdf/2207.08878v2.pdf
|
https://github.com/jingxiaoliu/bridge-damage-segmentation
| true | true | true |
pytorch
|
https://paperswithcode.com/paper/kold-korean-offensive-language-dataset
|
KOLD: Korean Offensive Language Dataset
|
2205.11315
|
https://arxiv.org/abs/2205.11315v2
|
https://arxiv.org/pdf/2205.11315v2.pdf
|
https://github.com/boychaboy/kold
| true | true | true |
none
|
https://paperswithcode.com/paper/wenet-2-0-more-productive-end-to-end-speech
|
WeNet 2.0: More Productive End-to-End Speech Recognition Toolkit
|
2203.15455
|
https://arxiv.org/abs/2203.15455v2
|
https://arxiv.org/pdf/2203.15455v2.pdf
|
https://github.com/wenet-e2e/wenet
| true | true | true |
pytorch
|
https://paperswithcode.com/paper/interpretable-semantic-photo-geolocalization
|
Interpretable Semantic Photo Geolocation
|
2104.14995
|
https://arxiv.org/abs/2104.14995v2
|
https://arxiv.org/pdf/2104.14995v2.pdf
|
https://github.com/jtheiner/semantic_geo_partitioning
| true | true | true |
pytorch
|
https://paperswithcode.com/paper/high-resolution-image-synthesis-with-latent
|
High-Resolution Image Synthesis with Latent Diffusion Models
|
2112.10752
|
https://arxiv.org/abs/2112.10752v2
|
https://arxiv.org/pdf/2112.10752v2.pdf
|
https://github.com/joanrod/ocr-vqgan
| false | false | true |
pytorch
|
https://paperswithcode.com/paper/taming-transformers-for-high-resolution-image
|
Taming Transformers for High-Resolution Image Synthesis
|
2012.09841
|
https://arxiv.org/abs/2012.09841v3
|
https://arxiv.org/pdf/2012.09841v3.pdf
|
https://github.com/joanrod/ocr-vqgan
| false | false | true |
pytorch
|
https://paperswithcode.com/paper/character-region-awareness-for-text-detection
|
Character Region Awareness for Text Detection
|
1904.01941
|
http://arxiv.org/abs/1904.01941v1
|
http://arxiv.org/pdf/1904.01941v1.pdf
|
https://github.com/joanrod/ocr-vqgan
| false | false | true |
pytorch
|
https://paperswithcode.com/paper/cryptanalyzing-an-image-encryption-algorithm-1
|
Cryptanalyzing an Image Encryption Algorithm Underpinned by 2D Lag-Complex Logistic Map
|
2208.06774
|
https://arxiv.org/abs/2208.06774v1
|
https://arxiv.org/pdf/2208.06774v1.pdf
|
https://github.com/chengqingli/mm-iealm
| true | true | false |
none
|
https://paperswithcode.com/paper/rethinking-image-mixture-for-unsupervised
|
Un-Mix: Rethinking Image Mixtures for Unsupervised Visual Representation Learning
|
2003.05438
|
https://arxiv.org/abs/2003.05438v5
|
https://arxiv.org/pdf/2003.05438v5.pdf
|
https://github.com/szq0214/Rethinking-Image-Mixture-for-Unsupervised-Learning
| true | true | true |
pytorch
|
https://paperswithcode.com/paper/tristereonet-a-trinocular-framework-for-multi
|
TriStereoNet: A Trinocular Framework for Multi-baseline Disparity Estimation
|
2111.12502
|
https://arxiv.org/abs/2111.12502v2
|
https://arxiv.org/pdf/2111.12502v2.pdf
|
https://github.com/cogsys-tuebingen/tristereonet
| true | true | true |
pytorch
|
https://paperswithcode.com/paper/mid-fusion-octree-based-object-level-multi
|
MID-Fusion: Octree-based Object-Level Multi-Instance Dynamic SLAM
|
1812.07976
|
http://arxiv.org/abs/1812.07976v4
|
http://arxiv.org/pdf/1812.07976v4.pdf
|
https://github.com/smartroboticslab/mid-fusion
| true | false | true |
tf
|
https://paperswithcode.com/paper/deep-graph-library-towards-efficient-and
|
Deep Graph Library: A Graph-Centric, Highly-Performant Package for Graph Neural Networks
|
1909.01315
|
https://arxiv.org/abs/1909.01315v2
|
https://arxiv.org/pdf/1909.01315v2.pdf
|
https://github.com/OweysMomenzada/Graph-Neural-Networks-for-effecient-Recommender-Systems
| false | false | true |
pytorch
|
https://paperswithcode.com/paper/self-guided-contrastive-learning-for-bert
|
Self-Guided Contrastive Learning for BERT Sentence Representations
|
2106.07345
|
https://arxiv.org/abs/2106.07345v1
|
https://arxiv.org/pdf/2106.07345v1.pdf
|
https://github.com/galsang/SG-BERT
| true | true | false |
pytorch
|
https://paperswithcode.com/paper/xnor-net-imagenet-classification-using-binary
|
XNOR-Net: ImageNet Classification Using Binary Convolutional Neural Networks
|
1603.05279
|
http://arxiv.org/abs/1603.05279v4
|
http://arxiv.org/pdf/1603.05279v4.pdf
|
https://github.com/pminhtam/xnor_conv_pytorch_extension
| false | false | false |
pytorch
|
https://paperswithcode.com/paper/fleet-policy-learning-via-weight-merging-and
|
Robot Fleet Learning via Policy Merging
|
2310.01362
|
https://arxiv.org/abs/2310.01362v3
|
https://arxiv.org/pdf/2310.01362v3.pdf
|
https://github.com/liruiw/fleet-tools
| true | true | true |
none
|
https://paperswithcode.com/paper/multiclass-sgcn-sparse-graph-based-trajectory
|
Multiclass-SGCN: Sparse Graph-based Trajectory Prediction with Agent Class Embedding
|
2206.15275
|
https://arxiv.org/abs/2206.15275v1
|
https://arxiv.org/pdf/2206.15275v1.pdf
|
https://github.com/carrotsniper/multiclass-sgcn
| true | true | true |
pytorch
|
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.