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/diffusion-based-environment-aware-trajectory
|
Diffusion-Based Environment-Aware Trajectory Prediction
|
2403.11643
|
https://arxiv.org/abs/2403.11643v1
|
https://arxiv.org/pdf/2403.11643v1.pdf
|
https://github.com/westny/dronalize
| false | false | true |
pytorch
|
https://paperswithcode.com/paper/mtp-go-graph-based-probabilistic-multi-agent
|
MTP-GO: Graph-Based Probabilistic Multi-Agent Trajectory Prediction with Neural ODEs
|
2302.00735
|
https://arxiv.org/abs/2302.00735v4
|
https://arxiv.org/pdf/2302.00735v4.pdf
|
https://github.com/westny/dronalize
| false | false | true |
pytorch
|
https://paperswithcode.com/paper/evaluation-of-differentially-constrained
|
Evaluation of Differentially Constrained Motion Models for Graph-Based Trajectory Prediction
|
2304.05116
|
https://arxiv.org/abs/2304.05116v2
|
https://arxiv.org/pdf/2304.05116v2.pdf
|
https://github.com/westny/dronalize
| false | false | true |
pytorch
|
https://paperswithcode.com/paper/instructrag-instructing-retrieval-augmented
|
InstructRAG: Instructing Retrieval-Augmented Generation via Self-Synthesized Rationales
|
2406.13629
|
https://arxiv.org/abs/2406.13629v2
|
https://arxiv.org/pdf/2406.13629v2.pdf
|
https://github.com/weizhepei/instructrag
| true | true | true |
pytorch
|
https://paperswithcode.com/paper/facial-landmark-points-detection-using
|
Facial Landmark Points Detection Using Knowledge Distillation-Based Neural Networks
|
2111.07047
|
https://arxiv.org/abs/2111.07047v1
|
https://arxiv.org/pdf/2111.07047v1.pdf
|
https://github.com/aliprf/kd-loss
| true | true | true |
tf
|
https://paperswithcode.com/paper/seeing-clearly-answering-incorrectly-a
|
Unveiling the Ignorance of MLLMs: Seeing Clearly, Answering Incorrectly
|
2406.10638
|
https://arxiv.org/abs/2406.10638v2
|
https://arxiv.org/pdf/2406.10638v2.pdf
|
https://github.com/baai-dcai/multimodal-robustness-benchmark
| true | true | true |
pytorch
|
https://paperswithcode.com/paper/a-diagnostic-tool-for-functional-causal
|
A Diagnostic Tool for Functional Causal Discovery
|
2406.07787
|
https://arxiv.org/abs/2406.07787v2
|
https://arxiv.org/pdf/2406.07787v2.pdf
|
https://github.com/shreyap18/causalDiagnose
| true | true | false |
none
|
https://paperswithcode.com/paper/efficient-probabilistic-modeling-of
|
Efficient Probabilistic Modeling of Crystallization at Mesoscopic Scale
|
2405.16608
|
https://arxiv.org/abs/2405.16608v1
|
https://arxiv.org/pdf/2405.16608v1.pdf
|
https://github.com/poltimmer/CGNE
| true | true | true |
pytorch
|
https://paperswithcode.com/paper/validated-error-bounds-for-pseudospectral
|
Validated error bounds for pseudospectral approximation of delay differential equations: unstable manifolds
|
2405.07727
|
https://arxiv.org/abs/2405.07727v1
|
https://arxiv.org/pdf/2405.07727v1.pdf
|
https://github.com/skepley/pseudospectral_DDE_CAP
| true | true | true |
none
|
https://paperswithcode.com/paper/goat-bench-a-benchmark-for-multi-modal
|
GOAT-Bench: A Benchmark for Multi-Modal Lifelong Navigation
|
2404.06609
|
https://arxiv.org/abs/2404.06609v1
|
https://arxiv.org/pdf/2404.06609v1.pdf
|
https://github.com/Ram81/goat-bench
| true | false | false |
pytorch
|
https://paperswithcode.com/paper/safety-fine-tuning-at-almost-no-cost-a
|
Safety Fine-Tuning at (Almost) No Cost: A Baseline for Vision Large Language Models
|
2402.02207
|
https://arxiv.org/abs/2402.02207v2
|
https://arxiv.org/pdf/2402.02207v2.pdf
|
https://github.com/ys-zong/vlguard
| true | true | true |
pytorch
|
https://paperswithcode.com/paper/howtocaption-prompting-llms-to-transform
|
HowToCaption: Prompting LLMs to Transform Video Annotations at Scale
|
2310.04900
|
https://arxiv.org/abs/2310.04900v2
|
https://arxiv.org/pdf/2310.04900v2.pdf
|
https://github.com/ninatu/howtocaption
| true | true | true |
pytorch
|
https://paperswithcode.com/paper/scalable-random-feature-latent-variable
|
Scalable Random Feature Latent Variable Models
|
2410.17700
|
https://arxiv.org/abs/2410.17700v1
|
https://arxiv.org/pdf/2410.17700v1.pdf
|
https://github.com/gwgundersen/rflvm
| true | true | false |
none
|
https://paperswithcode.com/paper/exploring-scalability-in-large-scale-time
|
Exploring Scalability in Large-Scale Time Series in DeepVATS framework
|
2408.04692
|
https://arxiv.org/abs/2408.04692v1
|
https://arxiv.org/pdf/2408.04692v1.pdf
|
https://github.com/vrodriguezf/deepvats
| true | true | false |
tf
|
https://paperswithcode.com/paper/one-shot-face-sketch-synthesis-in-the-wild
|
One-shot Face Sketch Synthesis in the Wild via Generative Diffusion Prior and Instruction Tuning
|
2506.15312
|
https://arxiv.org/abs/2506.15312v1
|
https://arxiv.org/pdf/2506.15312v1.pdf
|
https://github.com/hanwu3125/os-sketch
| true | true | false |
none
|
https://paperswithcode.com/paper/translating-mathematical-formula-images-to
|
Translating Math Formula Images to LaTeX Sequences Using Deep Neural Networks with Sequence-level Training
|
1908.11415
|
https://arxiv.org/abs/1908.11415v2
|
https://arxiv.org/pdf/1908.11415v2.pdf
|
https://github.com/pwc-1/Paper-9/tree/main/4/translating-math-formula-images
| false | false | false |
mindspore
|
https://paperswithcode.com/paper/contactdb-analyzing-and-predicting-grasp
|
ContactDB: Analyzing and Predicting Grasp Contact via Thermal Imaging
|
1904.06830
|
http://arxiv.org/abs/1904.06830v1
|
http://arxiv.org/pdf/1904.06830v1.pdf
|
https://github.com/samarth-robo/contactdb_prediction
| true | false | true |
pytorch
|
https://paperswithcode.com/paper/tc-kanrecon-high-quality-and-accelerated-mri
|
TC-KANRecon: High-Quality and Accelerated MRI Reconstruction via Adaptive KAN Mechanisms and Intelligent Feature Scaling
|
2408.05705
|
https://arxiv.org/abs/2408.05705v2
|
https://arxiv.org/pdf/2408.05705v2.pdf
|
https://github.com/lcbkmm/tc-kanrecon
| true | true | false |
pytorch
|
https://paperswithcode.com/paper/time-matters-examine-temporal-effects-on
|
Time Matters: Examine Temporal Effects on Biomedical Language Models
|
2407.17638
|
https://arxiv.org/abs/2407.17638v2
|
https://arxiv.org/pdf/2407.17638v2.pdf
|
https://github.com/trust-nlp/temporalassessment
| true | true | true |
tf
|
https://paperswithcode.com/paper/residual-inr-communication-efficient-on
|
Residual-INR: Communication Efficient On-Device Learning Using Implicit Neural Representation
|
2408.05617
|
https://arxiv.org/abs/2408.05617v3
|
https://arxiv.org/pdf/2408.05617v3.pdf
|
https://github.com/sharc-lab/residual-inr
| true | true | false |
pytorch
|
https://paperswithcode.com/paper/mplug-owl3-towards-long-image-sequence
|
mPLUG-Owl3: Towards Long Image-Sequence Understanding in Multi-Modal Large Language Models
|
2408.04840
|
https://arxiv.org/abs/2408.04840v2
|
https://arxiv.org/pdf/2408.04840v2.pdf
|
https://github.com/x-plug/mplug-owl
| true | true | false |
pytorch
|
https://paperswithcode.com/paper/shapley-pc-constraint-based-causal-structure
|
Shapley-PC: Constraint-based Causal Structure Learning with a Shapley Inspired Framework
|
2312.11582
|
https://arxiv.org/abs/2312.11582v3
|
https://arxiv.org/pdf/2312.11582v3.pdf
|
https://github.com/briziorusso/shapleypc
| true | true | false |
pytorch
|
https://paperswithcode.com/paper/finding-meaning-in-points-weakly-supervised
|
Finding Meaning in Points: Weakly Supervised Semantic Segmentation for Event Cameras
|
2407.11216
|
https://arxiv.org/abs/2407.11216v1
|
https://arxiv.org/pdf/2407.11216v1.pdf
|
https://github.com/chohoonhee/ev-wsss
| true | true | true |
pytorch
|
https://paperswithcode.com/paper/activegs-active-scene-reconstruction-using
|
ActiveGS: Active Scene Reconstruction Using Gaussian Splatting
|
2412.17769
|
https://arxiv.org/abs/2412.17769v2
|
https://arxiv.org/pdf/2412.17769v2.pdf
|
https://github.com/dmar-bonn/active-gs
| true | true | true |
jax
|
https://paperswithcode.com/paper/crab-cross-environment-agent-benchmark-for
|
CRAB: Cross-environment Agent Benchmark for Multimodal Language Model Agents
|
2407.01511
|
https://arxiv.org/abs/2407.01511v2
|
https://arxiv.org/pdf/2407.01511v2.pdf
|
https://github.com/camel-ai/crab
| true | true | true |
pytorch
|
https://paperswithcode.com/paper/forecasting-railway-ticket-demand-with-search
|
Forecasting railway ticket demand with search query open data
| null |
https://www.sciencedirect.com/science/article/pii/S1877050922016878?via%3Dihub
|
https://www.sciencedirect.com/science/article/pii/S1877050922016878?via%3Dihub
|
https://github.com/AlgoMathITMO/Forecasting-railway-ticket-demand-with-search-query
| false | false | false |
none
|
https://paperswithcode.com/paper/mic-drop-on-estimating-the-size-of-sub-mm
|
'Mic drop': on estimating the size of sub-mm droplets using a simple condenser microphone
|
2506.19782
|
https://arxiv.org/abs/2506.19782v1
|
https://arxiv.org/pdf/2506.19782v1.pdf
|
https://github.com/avof/mic-drop
| true | true | false |
pytorch
|
https://paperswithcode.com/paper/assessing-student-s-dynamic-knowledge-state
|
Assessing Student's Dynamic Knowledge State by Exploring the Question Difficulty Effect
| null |
https://dl.acm.org/doi/abs/10.1145/3477495.3531939
|
https://dl.acm.org/doi/abs/10.1145/3477495.3531939
|
https://github.com/shshen-closer/DIMKT
| false | true | false |
tf
|
https://paperswithcode.com/paper/heuristic-dropout-an-efficient-regularization
|
Heuristic Dropout: An Efficient Regularization Method for Medical Image Segmentation Models
| null |
https://ieeexplore.ieee.org/abstract/document/9747409
|
https://ieeexplore.ieee.org/abstract/document/9747409
|
https://github.com/MindCode-4/code-7/tree/main/HeuristicDropout
| false | false | false |
mindspore
|
https://paperswithcode.com/paper/navigating-the-effect-of-parametrization-for
|
Navigating the Effect of Parametrization for Dimensionality Reduction
|
2411.15894
|
https://arxiv.org/abs/2411.15894v1
|
https://arxiv.org/pdf/2411.15894v1.pdf
|
https://github.com/hyhuang00/paramrepulsor
| true | true | false |
pytorch
|
https://paperswithcode.com/paper/meeg-and-at-dgnn-advancing-eeg-emotion
|
MEEG and AT-DGNN: Improving EEG Emotion Recognition with Music Introducing and Graph-based Learning
|
2407.05550
|
https://arxiv.org/abs/2407.05550v4
|
https://arxiv.org/pdf/2407.05550v4.pdf
|
https://github.com/xmh1011/at-dgnn
| true | true | true |
pytorch
|
https://paperswithcode.com/paper/toward-self-improvement-of-llms-via
|
Toward Self-Improvement of LLMs via Imagination, Searching, and Criticizing
|
2404.12253
|
https://arxiv.org/abs/2404.12253v2
|
https://arxiv.org/pdf/2404.12253v2.pdf
|
https://github.com/yetianjhu/alphallm
| true | true | false |
pytorch
|
https://paperswithcode.com/paper/community-research-earth-digital-intelligence
|
Community Research Earth Digital Intelligence Twin (CREDIT)
|
2411.07814
|
https://arxiv.org/abs/2411.07814v1
|
https://arxiv.org/pdf/2411.07814v1.pdf
|
https://github.com/ncar/credit-arxiv
| true | true | true |
none
|
https://paperswithcode.com/paper/quantile-deep-learning-models-for-multi-step
|
Quantile deep learning models for multi-step ahead time series prediction
|
2411.15674
|
https://arxiv.org/abs/2411.15674v1
|
https://arxiv.org/pdf/2411.15674v1.pdf
|
https://github.com/sydney-machine-learning/quantiledeeplearning
| true | true | false |
none
|
https://paperswithcode.com/paper/pint-a-modern-software-package-for-pulsar
|
PINT: A Modern Software Package for Pulsar Timing
|
2012.00074
|
https://arxiv.org/abs/2012.00074v1
|
https://arxiv.org/pdf/2012.00074v1.pdf
|
https://github.com/nanograv/pint
| true | true | true |
none
|
https://paperswithcode.com/paper/factor-exposure-heterogeneity-in-green-and
|
Factor Exposure Heterogeneity in Green and Brown Stocks
|
2302.11729
|
https://arxiv.org/abs/2302.11729v2
|
https://arxiv.org/pdf/2302.11729v2.pdf
|
https://github.com/ardiad/peerperformance
| true | true | false |
none
|
https://paperswithcode.com/paper/how-easy-is-it-for-investment-managers-to
|
How easy is it for investment managers to deploy their talent in green and brown stocks?
|
2201.05709
|
https://arxiv.org/abs/2201.05709v2
|
https://arxiv.org/pdf/2201.05709v2.pdf
|
https://github.com/ardiad/peerperformance
| true | true | false |
none
|
https://paperswithcode.com/paper/nonintrusive-model-order-reduction-for
|
Learning Stochastic Reduced Models from Data: A Nonintrusive Approach
|
2407.05724
|
https://arxiv.org/abs/2407.05724v3
|
https://arxiv.org/pdf/2407.05724v3.pdf
|
https://github.com/jmnicolaus/operatorinference_for_sdes
| true | true | true |
none
|
https://paperswithcode.com/paper/the-ldbc-social-network-benchmark
|
The LDBC Social Network Benchmark
|
2001.02299
|
https://arxiv.org/abs/2001.02299v9
|
https://arxiv.org/pdf/2001.02299v9.pdf
|
https://github.com/ldbc/ldbc_snb_interactive_impls
| true | true | true |
none
|
https://paperswithcode.com/paper/regnlp-in-action-facilitating-compliance
|
RIRAG: Regulatory Information Retrieval and Answer Generation
|
2409.05677
|
https://arxiv.org/abs/2409.05677v2
|
https://arxiv.org/pdf/2409.05677v2.pdf
|
https://github.com/regnlp/obliqadataset
| true | true | true |
none
|
https://paperswithcode.com/paper/replay-a-recommendation-framework-for
|
RePlay: a Recommendation Framework for Experimentation and Production Use
|
2409.07272
|
https://arxiv.org/abs/2409.07272v3
|
https://arxiv.org/pdf/2409.07272v3.pdf
|
https://github.com/sb-ai-lab/RePlay
| true | true | true |
pytorch
|
https://paperswithcode.com/paper/sowa-adapting-hierarchical-frozen-window-self
|
SOWA: Adapting Hierarchical Frozen Window Self-Attention to Visual-Language Models for Better Anomaly Detection
|
2407.03634
|
https://arxiv.org/abs/2407.03634v4
|
https://arxiv.org/pdf/2407.03634v4.pdf
|
https://github.com/huzongxiang/sowa
| true | false | true |
pytorch
|
https://paperswithcode.com/paper/vg-tvp-multimodal-procedural-planning-via
|
VG-TVP: Multimodal Procedural Planning via Visually Grounded Text-Video Prompting
|
2412.11621
|
https://arxiv.org/abs/2412.11621v1
|
https://arxiv.org/pdf/2412.11621v1.pdf
|
https://github.com/mfurkanilaslan/vg-tvp
| true | true | true |
none
|
https://paperswithcode.com/paper/leveraging-the-doppler-effect-for-channel
|
Leveraging the Doppler Effect for Channel Charting
|
2404.09620
|
https://arxiv.org/abs/2404.09620v1
|
https://arxiv.org/pdf/2404.09620v1.pdf
|
https://github.com/jeija/doppler-effect-channelcharting
| true | true | true |
tf
|
https://paperswithcode.com/paper/direct-discriminative-optimization-your-1
|
Direct Discriminative Optimization: Your Likelihood-Based Visual Generative Model is Secretly a GAN Discriminator
|
2503.01103
|
https://arxiv.org/abs/2503.01103v2
|
https://arxiv.org/pdf/2503.01103v2.pdf
|
https://github.com/nvlabs/ddo
| true | true | true |
pytorch
|
https://paperswithcode.com/paper/prompto-an-open-source-library-for
|
Prompto: An open source library for asynchronous querying of LLM endpoints
|
2408.11847
|
https://arxiv.org/abs/2408.11847v2
|
https://arxiv.org/pdf/2408.11847v2.pdf
|
https://github.com/alan-turing-institute/prompto
| true | true | true |
none
|
https://paperswithcode.com/paper/on-the-design-and-analysis-of-llm-based
|
On the Design and Analysis of LLM-Based Algorithms
|
2407.14788
|
https://arxiv.org/abs/2407.14788v2
|
https://arxiv.org/pdf/2407.14788v2.pdf
|
https://github.com/modelscope/agentscope
| true | true | true |
none
|
https://paperswithcode.com/paper/assumption-lean-and-data-adaptive-post
|
Assumption-Lean and Data-Adaptive Post-Prediction Inference
|
2311.14220
|
https://arxiv.org/abs/2311.14220v4
|
https://arxiv.org/pdf/2311.14220v4.pdf
|
https://github.com/qlu-lab/popinf
| true | true | true |
none
|
https://paperswithcode.com/paper/very-large-scale-multi-agent-simulation-in
|
Very Large-Scale Multi-Agent Simulation in AgentScope
|
2407.17789
|
https://arxiv.org/abs/2407.17789v2
|
https://arxiv.org/pdf/2407.17789v2.pdf
|
https://github.com/modelscope/agentscope
| true | true | true |
none
|
https://paperswithcode.com/paper/contactless-cardiac-pulse-monitoring-using
|
Contactless Cardiac Pulse Monitoring Using Event Cameras
|
2505.09529
|
https://arxiv.org/abs/2505.09529v2
|
https://arxiv.org/pdf/2505.09529v2.pdf
|
https://github.com/c3imaging/contactless_cardiac_pulse_monitoring_using_event_cameras
| true | true | false |
pytorch
|
https://paperswithcode.com/paper/sauc-sparsity-aware-uncertainty-calibration
|
SAUC: Sparsity-Aware Uncertainty Calibration for Spatiotemporal Prediction with Graph Neural Networks
|
2409.08766
|
https://arxiv.org/abs/2409.08766v1
|
https://arxiv.org/pdf/2409.08766v1.pdf
|
https://github.com/AnonymousSAUC/SAUC
| true | false | false |
pytorch
|
https://paperswithcode.com/paper/modeling-of-terrain-deformation-by-a-grouser
|
Modeling of Terrain Deformation by a Grouser Wheel for Lunar Rover Simulation
|
2408.13468
|
https://arxiv.org/abs/2408.13468v1
|
https://arxiv.org/pdf/2408.13468v1.pdf
|
https://github.com/antoinerichard/lunarsim
| false | false | true |
none
|
https://paperswithcode.com/paper/neural-rendering-for-stereo-3d-reconstruction
|
Neural Rendering for Stereo 3D Reconstruction of Deformable Tissues in Robotic Surgery
|
2206.15255
|
https://arxiv.org/abs/2206.15255v1
|
https://arxiv.org/pdf/2206.15255v1.pdf
|
https://github.com/CUHK-AIM-Group/LGS
| false | false | true |
pytorch
|
https://paperswithcode.com/paper/alpapico-extraction-of-pico-frames-from
|
AlpaPICO: Extraction of PICO Frames from Clinical Trial Documents Using LLMs
|
2409.09704
|
https://arxiv.org/abs/2409.09704v1
|
https://arxiv.org/pdf/2409.09704v1.pdf
|
https://github.com/shrimonmuke0202/alpapico
| true | true | false |
pytorch
|
https://paperswithcode.com/paper/diffusion-models-for-stronger-face-morphing
|
Leveraging Diffusion For Strong and High Quality Face Morphing Attacks
|
2301.04218
|
https://arxiv.org/abs/2301.04218v4
|
https://arxiv.org/pdf/2301.04218v4.pdf
|
https://github.com/zblasingame/DiM
| true | false | true |
pytorch
|
https://paperswithcode.com/paper/bias-reduction-in-matched-observational
|
Bias Mitigation in Matched Observational Studies with Continuous Treatments: Calipered Non-Bipartite Matching and Bias-Corrected Estimation and Inference
|
2409.11701
|
https://arxiv.org/abs/2409.11701v2
|
https://arxiv.org/pdf/2409.11701v2.pdf
|
https://github.com/anthonyfraziercsu/mitigating-bias-matched-observational-studies
| true | true | false |
none
|
https://paperswithcode.com/paper/the-leray-transform-distinguished-measures
|
The Leray transform: distinguished measures, symmetries and polygamma inequalities
|
2401.17490
|
https://arxiv.org/abs/2401.17490v3
|
https://arxiv.org/pdf/2401.17490v3.pdf
|
https://github.com/ledholm/leray-measures-2024-mathematica-nb
| true | true | false |
none
|
https://paperswithcode.com/paper/docvxqa-context-aware-visual-explanations-for
|
DocVXQA: Context-Aware Visual Explanations for Document Question Answering
|
2505.07496
|
https://arxiv.org/abs/2505.07496v1
|
https://arxiv.org/pdf/2505.07496v1.pdf
|
https://github.com/dali92002/docvxqa
| true | true | false |
pytorch
|
https://paperswithcode.com/paper/the-factuality-of-large-language-models-in
|
The Factuality of Large Language Models in the Legal Domain
|
2409.11798
|
https://arxiv.org/abs/2409.11798v1
|
https://arxiv.org/pdf/2409.11798v1.pdf
|
https://github.com/rajjaa/lexfact
| true | true | false |
pytorch
|
https://paperswithcode.com/paper/synchronization-of-wave-propelled-capillary
|
Synchronization of wave-propelled capillary spinners
|
2409.06652
|
https://arxiv.org/abs/2409.06652v2
|
https://arxiv.org/pdf/2409.06652v2.pdf
|
https://github.com/harrislab-brown/syncspinners
| true | true | false |
none
|
https://paperswithcode.com/paper/utilizing-description-logics-for-global
|
Utilizing Description Logics for Global Explanations of Heterogeneous Graph Neural Networks
|
2405.12654
|
https://arxiv.org/abs/2405.12654v1
|
https://arxiv.org/pdf/2405.12654v1.pdf
|
https://github.com/ds-jrg/xgnn-dl
| true | true | true |
pytorch
|
https://paperswithcode.com/paper/robostack-using-the-robot-operating-system
|
A RoboStack Tutorial: Using the Robot Operating System Alongside the Conda and Jupyter Data Science Ecosystems
|
2104.12910
|
https://arxiv.org/abs/2104.12910v3
|
https://arxiv.org/pdf/2104.12910v3.pdf
|
https://github.com/mbatc/ros-humble
| false | false | true |
pytorch
|
https://paperswithcode.com/paper/ssr-speech-towards-stable-safe-and-robust
|
SSR-Speech: Towards Stable, Safe and Robust Zero-shot Text-based Speech Editing and Synthesis
|
2409.07556
|
https://arxiv.org/abs/2409.07556v2
|
https://arxiv.org/pdf/2409.07556v2.pdf
|
https://github.com/WangHelin1997/SSR-Speech
| true | false | true |
pytorch
|
https://paperswithcode.com/paper/improving-consistency-in-large-language
|
Improving Consistency in Large Language Models through Chain of Guidance
|
2502.15924
|
https://arxiv.org/abs/2502.15924v1
|
https://arxiv.org/pdf/2502.15924v1.pdf
|
https://github.com/vijilAI/chain_of_guidance
| true | false | false |
pytorch
|
https://paperswithcode.com/paper/towards-deep-generation-of-guided-wave
|
Towards deep generation of guided wave representations for composite materials
|
2212.06365
|
https://arxiv.org/abs/2212.06365v1
|
https://arxiv.org/pdf/2212.06365v1.pdf
|
https://github.com/mahindrautela/deepgenerator_compositematerialgwrepresentations
| true | true | true |
tf
|
https://paperswithcode.com/paper/prompt-agnostic-adversarial-perturbation-for
|
Prompt-Agnostic Adversarial Perturbation for Customized Diffusion Models
|
2408.10571
|
https://arxiv.org/abs/2408.10571v4
|
https://arxiv.org/pdf/2408.10571v4.pdf
|
https://github.com/vancyland/prompt-agnostic-adversarial-perturbation-for-customized-diffusion-models.github.io
| true | true | true |
pytorch
|
https://paperswithcode.com/paper/demystifying-large-language-models-for
|
Demystifying Large Language Models for Medicine: A Primer
|
2410.18856
|
https://arxiv.org/abs/2410.18856v3
|
https://arxiv.org/pdf/2410.18856v3.pdf
|
https://github.com/ncbi-nlp/llm-medicine-primer
| true | true | false |
none
|
https://paperswithcode.com/paper/does-differential-privacy-impact-bias-in
|
Does Differential Privacy Impact Bias in Pretrained NLP Models?
|
2410.18749
|
https://arxiv.org/abs/2410.18749v1
|
https://arxiv.org/pdf/2410.18749v1.pdf
|
https://github.com/khairulislam/dp-on-nlp-bias
| true | true | false |
pytorch
|
https://paperswithcode.com/paper/in-context-contrastive-learning-for-event
|
In-context Contrastive Learning for Event Causality Identification
|
2405.10512
|
https://arxiv.org/abs/2405.10512v2
|
https://arxiv.org/pdf/2405.10512v2.pdf
|
https://github.com/ChaoLiang-HUST/ICCL
| true | false | false |
pytorch
|
https://paperswithcode.com/paper/self-supervised-learning-for-time-series-a
|
Self-Supervised Learning for Time Series: A Review & Critique of FITS
|
2410.18318
|
https://arxiv.org/abs/2410.18318v1
|
https://arxiv.org/pdf/2410.18318v1.pdf
|
https://github.com/thorhojhus/ssl_fts
| true | true | false |
pytorch
|
https://paperswithcode.com/paper/exploiting-interpretable-capabilities-with
|
Exploiting Interpretable Capabilities with Concept-Enhanced Diffusion and Prototype Networks
|
2410.18705
|
https://arxiv.org/abs/2410.18705v2
|
https://arxiv.org/pdf/2410.18705v2.pdf
|
https://github.com/acarballocastro/ConceptEnhanced
| true | false | false |
pytorch
|
https://paperswithcode.com/paper/models-are-codes-towards-measuring-malicious
|
Models Are Codes: Towards Measuring Malicious Code Poisoning Attacks on Pre-trained Model Hubs
|
2409.09368
|
https://arxiv.org/abs/2409.09368v1
|
https://arxiv.org/pdf/2409.09368v1.pdf
|
https://github.com/security-pride/MalHug
| true | false | false |
tf
|
https://paperswithcode.com/paper/a-logical-fallacy-informed-framework-for
|
A Logical Fallacy-Informed Framework for Argument Generation
|
2408.03618
|
https://arxiv.org/abs/2408.03618v4
|
https://arxiv.org/pdf/2408.03618v4.pdf
|
https://github.com/lucamouchel/Logical-Fallacies
| true | true | false |
pytorch
|
https://paperswithcode.com/paper/naturalspeech-2-latent-diffusion-models-are
|
NaturalSpeech 2: Latent Diffusion Models are Natural and Zero-Shot Speech and Singing Synthesizers
|
2304.09116
|
https://arxiv.org/abs/2304.09116v3
|
https://arxiv.org/pdf/2304.09116v3.pdf
|
https://github.com/adelacvg/ns2vc
| false | false | true |
pytorch
|
https://paperswithcode.com/paper/on-device-collaborative-language-modeling-via
|
On-Device Collaborative Language Modeling via a Mixture of Generalists and Specialists
|
2409.13931
|
https://arxiv.org/abs/2409.13931v2
|
https://arxiv.org/pdf/2409.13931v2.pdf
|
https://github.com/epfml/comigs
| true | true | true |
pytorch
|
https://paperswithcode.com/paper/attention-score-is-not-all-you-need-for-token
|
Attention Score is not All You Need for Token Importance Indicator in KV Cache Reduction: Value Also Matters
|
2406.12335
|
https://arxiv.org/abs/2406.12335v2
|
https://arxiv.org/pdf/2406.12335v2.pdf
|
https://github.com/guozhiyu/vatp
| true | true | true |
pytorch
|
https://paperswithcode.com/paper/stgformer-efficient-spatiotemporal-graph
|
STGformer: Efficient Spatiotemporal Graph Transformer for Traffic Forecasting
|
2410.00385
|
https://arxiv.org/abs/2410.00385v2
|
https://arxiv.org/pdf/2410.00385v2.pdf
|
https://github.com/dreamzz5/stgformer
| true | true | false |
pytorch
|
https://paperswithcode.com/paper/a-spectral-framework-for-tracking-communities
|
A Spectral Framework for Tracking Communities in Evolving Networks
|
2412.07378
|
https://arxiv.org/abs/2412.07378v1
|
https://arxiv.org/pdf/2412.07378v1.pdf
|
https://github.com/jacobh140/spectral-dcd
| true | false | false |
none
|
https://paperswithcode.com/paper/vleu-a-method-for-automatic-evaluation-for
|
VLEU: a Method for Automatic Evaluation for Generalizability of Text-to-Image Models
|
2409.14704
|
https://arxiv.org/abs/2409.14704v2
|
https://arxiv.org/pdf/2409.14704v2.pdf
|
https://github.com/mio7690/VLEU
| true | false | true |
pytorch
|
https://paperswithcode.com/paper/diffsf-diffusion-models-for-scene-flow
|
DiffSF: Diffusion Models for Scene Flow Estimation
|
2403.05327
|
https://arxiv.org/abs/2403.05327v3
|
https://arxiv.org/pdf/2403.05327v3.pdf
|
https://github.com/zhangyushan3/diffsf
| true | true | true |
pytorch
|
https://paperswithcode.com/paper/advantage-guided-distillation-for-preference
|
Advantage-Guided Distillation for Preference Alignment in Small Language Models
|
2502.17927
|
https://arxiv.org/abs/2502.17927v1
|
https://arxiv.org/pdf/2502.17927v1.pdf
|
https://github.com/slit-ai/adpa
| true | true | false |
pytorch
|
https://paperswithcode.com/paper/nuscenes-a-multimodal-dataset-for-autonomous
|
nuScenes: A multimodal dataset for autonomous driving
|
1903.11027
|
https://arxiv.org/abs/1903.11027v5
|
https://arxiv.org/pdf/1903.11027v5.pdf
|
https://github.com/Ggs1mida/Awesome-DataFusion
| false | false | true |
none
|
https://paperswithcode.com/paper/eulerian-simulation-of-complex-suspensions
|
Eulerian simulation of complex suspensions and biolocomotion in three dimensions
|
2104.00095
|
https://arxiv.org/abs/2104.00095v1
|
https://arxiv.org/pdf/2104.00095v1.pdf
|
https://github.com/ylunalin/rmt3D
| true | true | true |
none
|
https://paperswithcode.com/paper/re-assembling-the-past-the-repair-dataset-and
|
Re-assembling the past: The RePAIR dataset and benchmark for real world 2D and 3D puzzle solving
|
2410.24010
|
https://arxiv.org/abs/2410.24010v2
|
https://arxiv.org/pdf/2410.24010v2.pdf
|
https://github.com/RePAIRProject/repair_ground_truth
| false | false | false |
none
|
https://paperswithcode.com/paper/floonoc-a-645-gbps-link-0-15-pj-b-hop-open
|
FlooNoC: A 645 Gbps/link 0.15 pJ/B/hop Open-Source NoC with Wide Physical Links and End-to-End AXI4 Parallel Multi-Stream Support
|
2409.17606
|
https://arxiv.org/abs/2409.17606v2
|
https://arxiv.org/pdf/2409.17606v2.pdf
|
https://github.com/pulp-platform/floonoc
| true | false | true |
none
|
https://paperswithcode.com/paper/an-analytically-tractable-marked-power
|
An Analytically Tractable Marked Power Spectrum
|
2409.17133
|
https://arxiv.org/abs/2409.17133v2
|
https://arxiv.org/pdf/2409.17133v2.pdf
|
https://github.com/HarukiEbina/markedPS
| true | false | true |
none
|
https://paperswithcode.com/paper/candoit-causal-discovery-with-observational
|
CAnDOIT: Causal Discovery with Observational and Interventional Data from Time-Series
|
2410.02844
|
https://arxiv.org/abs/2410.02844v3
|
https://arxiv.org/pdf/2410.02844v3.pdf
|
https://github.com/lcastri/causalflow
| true | true | true |
pytorch
|
https://paperswithcode.com/paper/cheating-automatic-llm-benchmarks-null-models
|
Cheating Automatic LLM Benchmarks: Null Models Achieve High Win Rates
|
2410.07137
|
https://arxiv.org/abs/2410.07137v2
|
https://arxiv.org/pdf/2410.07137v2.pdf
|
https://github.com/sail-sg/Cheating-LLM-Benchmarks
| true | true | true |
none
|
https://paperswithcode.com/paper/dana-domain-aware-neurosymbolic-agents-for
|
DANA: Domain-Aware Neurosymbolic Agents for Consistency and Accuracy
|
2410.02823
|
https://arxiv.org/abs/2410.02823v1
|
https://arxiv.org/pdf/2410.02823v1.pdf
|
https://github.com/aitomatic/openssa
| false | false | true |
none
|
https://paperswithcode.com/paper/hlv-1k-a-large-scale-hour-long-video
|
HLV-1K: A Large-scale Hour-Long Video Benchmark for Time-Specific Long Video Understanding
|
2501.01645
|
https://arxiv.org/abs/2501.01645v3
|
https://arxiv.org/pdf/2501.01645v3.pdf
|
https://github.com/vincent-zhq/hlv-1k
| true | true | false |
none
|
https://paperswithcode.com/paper/fed-biomed-open-transparent-and-trusted
|
Fed-BioMed: Open, Transparent and Trusted Federated Learning for Real-world Healthcare Applications
|
2304.12012
|
https://arxiv.org/abs/2304.12012v1
|
https://arxiv.org/pdf/2304.12012v1.pdf
|
https://github.com/fedbiomed/fedbiomed
| false | false | true |
pytorch
|
https://paperswithcode.com/paper/t2i-fineeval-fine-grained-compositional
|
T2I-FineEval: Fine-Grained Compositional Metric for Text-to-Image Evaluation
|
2503.11481
|
https://arxiv.org/abs/2503.11481v1
|
https://arxiv.org/pdf/2503.11481v1.pdf
|
https://github.com/hadi-hosseini/t2i-fineeval
| false | true | true |
pytorch
|
https://paperswithcode.com/paper/meshgpt-generating-triangle-meshes-with
|
MeshGPT: Generating Triangle Meshes with Decoder-Only Transformers
|
2311.15475
|
https://arxiv.org/abs/2311.15475v1
|
https://arxiv.org/pdf/2311.15475v1.pdf
|
https://github.com/lucidrains/meshgpt-pytorch
| false | false | true |
pytorch
|
https://paperswithcode.com/paper/eliminating-oversaturation-and-artifacts-of
|
Eliminating Oversaturation and Artifacts of High Guidance Scales in Diffusion Models
|
2410.02416
|
https://arxiv.org/abs/2410.02416v1
|
https://arxiv.org/pdf/2410.02416v1.pdf
|
https://github.com/lucidrains/meshgpt-pytorch
| false | false | true |
pytorch
|
https://paperswithcode.com/paper/qiskit-pulse-programming-quantum-computers
|
Qiskit Pulse: Programming Quantum Computers Through the Cloud with Pulses
|
2004.06755
|
http://arxiv.org/abs/2004.06755v1
|
http://arxiv.org/pdf/2004.06755v1.pdf
|
https://github.com/kashish0405/Gate-Optimisation
| false | false | true |
none
|
https://paperswithcode.com/paper/optimized-compilation-of-aggregated
|
Optimized Compilation of Aggregated Instructions for Realistic Quantum Computers
|
1902.01474
|
http://arxiv.org/abs/1902.01474v2
|
http://arxiv.org/pdf/1902.01474v2.pdf
|
https://github.com/kashish0405/Gate-Optimisation
| false | false | true |
none
|
https://paperswithcode.com/paper/optimized-quantum-compilation-for-near-term
|
Optimized Quantum Compilation for Near-Term Algorithms with OpenPulse
|
2004.11205
|
https://arxiv.org/abs/2004.11205v2
|
https://arxiv.org/pdf/2004.11205v2.pdf
|
https://github.com/kashish0405/Gate-Optimisation
| false | false | true |
none
|
https://paperswithcode.com/paper/tesseract-a-search-based-decoder-for-quantum
|
Tesseract: A Search-Based Decoder for Quantum Error Correction
|
2503.10988
|
https://arxiv.org/abs/2503.10988v1
|
https://arxiv.org/pdf/2503.10988v1.pdf
|
https://github.com/quantumlib/tesseract-decoder
| true | true | true |
none
|
https://paperswithcode.com/paper/collaborative-text-editing-with-eg-walker
|
Collaborative Text Editing with Eg-walker: Better, Faster, Smaller
|
2409.14252
|
https://arxiv.org/abs/2409.14252v1
|
https://arxiv.org/pdf/2409.14252v1.pdf
|
https://github.com/josephg/eg-walker-reference
| true | true | true |
none
|
https://paperswithcode.com/paper/progressive-neural-compression-for-adaptive
|
Progressive Neural Compression for Adaptive Image Offloading under Timing Constraints
|
2310.05306
|
https://arxiv.org/abs/2310.05306v1
|
https://arxiv.org/pdf/2310.05306v1.pdf
|
https://github.com/rickywrq/Progressive-Neural-Compression
| true | false | true |
tf
|
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.