Papers

New on arXiv

Mon 10 Feb 2025

410 papers

Deciphering Functions of Neurons in Vision-Language ModelsMultimodal LLMs / vision-language models → VLM benchmarks and visual reasoning
Towards Copyright Protection for Knowledge Bases of Retrieval-augmented Language Models via ReasoningLLM reasoning → Knowledge- and tool-augmented reasoning +2
Preference Alignment on Diffusion Model: A Comprehensive Survey for Image Generation and EditingDiffusion models → Diffusion and flow policies for robotics +1
Implicit Language Models are RNNs: Balancing Parallelization and ExpressivityLLM reasoning → Test-time compute and efficient thinking
Cross-platform Learning-based Fault Tolerant Surfacing Controller for Underwater RobotsDeep reinforcement learning methods → RL for robot navigation and UAV control
Interactive Data Harmonization with LLM Agents: Opportunities and ChallengesLLM agents and tool use → Web, search, and research agents
Online Scheduling for LLM Inference with KV Cache ConstraintsEfficient LLM inference and serving → KV cache optimization
Generative Distribution Prediction: A Unified Approach to Multimodal LearningDiffusion models → Image generation, text-to-image and editing
Evaluating the Systematic Reasoning Abilities of Large Language Models through Graph ColoringLLM reasoning → Knowledge- and tool-augmented reasoning
DROP: Poison Dilution via Knowledge Distillation for Federated LearningLLM safety and security → Federated unlearning and poisoning +1
From Image to Video: An Empirical Study of Diffusion RepresentationsDiffusion models → Video generation
Outsourced diffusion sampling: Efficient posterior inference in latent spaces of generative modelsDiffusion models → Diffusion foundations and sampling
Conditional diffusion model with spatial attention and latent embedding for medical image segmentationDiffusion models → Medical imaging and segmentation
A view on learning robust goal-conditioned value functions: Interplay between RL and MPCDeep reinforcement learning methods → RL for autonomous driving and traffic
SyncMind: Measuring Agent Out-of-Sync Recovery in Collaborative Software EngineeringLLM agents and tool use → Software engineering coding agents
Position: Episodic Memory is the Missing Piece for Long-Term LLM AgentsLLM agents and tool use → Agent memory systems
Intelligent Offloading in Vehicular Edge Computing: A Comprehensive Review of Deep Reinforcement Learning Approaches and ArchitecturesDeep reinforcement learning methods → RL for energy, networks and operations
GAS: Generative Avatar Synthesis from a Single ImageDiffusion models → 3D, point cloud and scene generation
Do Attention Heads Compete or Cooperate during Counting?Mechanistic interpretability → Circuits and mechanistic analysis
Select before Act: Spatially Decoupled Action Repetition for Continuous ControlDeep reinforcement learning methods → Deep RL algorithms and representations
Krum Federated Chain (KFC): Using blockchain to defend against adversarial attacks in Federated LearningLLM safety and security → Federated unlearning and poisoning
EVEv2: Improved Baselines for Encoder-Free Vision-Language ModelsMultimodal LLMs / vision-language models → VLM benchmarks and visual reasoning
Visual Agentic AI for Spatial Reasoning with a Dynamic APILLM reasoning → Spatial and embodied reasoning +1
Lumina-Video: Efficient and Flexible Video Generation with Multi-scale Next-DiTDiffusion models → Video generation
Exploring the Limit of Outcome Reward for Learning Mathematical ReasoningLLM reasoning → Reinforcement learning for reasoning +1
InSTA: Towards Internet-Scale Training For AgentsLLM agents and tool use → Web, search, and research agents
On the Emergence of Thinking in LLMs I: Searching for the Right IntuitionLLM reasoning → Reinforcement learning for reasoning
ReasonFlux: Hierarchical LLM Reasoning via Scaling Thought TemplatesLLM reasoning → Test-time compute and efficient thinking
Train for the Worst, Plan for the Best: Understanding Token Ordering in Masked DiffusionsDiffusion models → Discrete and text diffusion
History-Guided Video DiffusionDiffusion models → Diffusion foundations and sampling
Rationalization Models for Text-to-SQLLLM reasoning → Knowledge- and tool-augmented reasoning
Interpretable and Testable Vision Features via Sparse AutoencodersMechanistic interpretability → Interpreting vision and multimodal models
Resurrecting saturated LLM benchmarks with adversarial encodingLLM reasoning → Reasoning benchmarks and evaluation
VersaPRM: Multi-Domain Process Reward Model via Synthetic Reasoning DataLLM reasoning → Reinforcement learning for reasoning +1
Can 1B LLM Surpass 405B LLM? Rethinking Compute-Optimal Test-Time ScalingLLM reasoning → Test-time compute and efficient thinking
Who Taught You That? Tracing Teachers in Model DistillationKnowledge distillation → Knowledge distillation for LLMs
In-Context Learning (and Unlearning) of Length BiasesLLM safety and security → Machine unlearning
Transparent NLP: Using RAG and LLM Alignment for Privacy Q&ARetrieval-augmented generation → Privacy and security of RAG +1
MoETuner: Optimized Mixture of Expert Serving with Balanced Expert Placement and Token RoutingMixture-of-experts → Efficient MoE training and inference systems
Conformal Predictions for Human Action Recognition with Vision-Language ModelsMultimodal LLMs / vision-language models → Embodied and 3D vision-language models
Unleashing the Potential of Pre-Trained Diffusion Models for Generalizable Person Re-IdentificationDiffusion models → Image generation, text-to-image and editing
TripoSG: High-Fidelity 3D Shape Synthesis using Large-Scale Rectified Flow ModelsDiffusion models → 3D, point cloud and scene generation
MaterialFusion: High-Quality, Zero-Shot, and Controllable Material Transfer with Diffusion ModelsDiffusion models → 3D, point cloud and scene generation
Hephaestus: Improving Fundamental Agent Capabilities of Large Language Models through Continual Pre-TrainingLLM agents and tool use → General LLM agent frameworks and benchmarks
Predictive Red Teaming: Breaking Policies Without Breaking RobotsOffline RL and imitation learning from data → Robot manipulation imitation learning +1
Membership Inference Risks in Quantized Models: A Theoretical and Empirical StudyLLM safety and security → Membership inference attacks
Large Language Models Meet Symbolic Provers for Logical Reasoning EvaluationLLM reasoning → Reasoning benchmarks and evaluation
Diffusion Models for Computational Neuroimaging: A SurveyDiffusion models → Medical imaging and segmentation
Ignore the KL Penalty! Boosting Exploration on Critical Tokens to Enhance RL Fine-TuningLLM reasoning → Reinforcement learning for reasoning
CustomVideoX: 3D Reference Attention Driven Dynamic Adaptation for Zero-Shot Customized Video Diffusion TransformersDiffusion models → Video generation
Model-Based Offline Reinforcement Learning with Reliability-Guaranteed Sequence ModelingOffline RL and imitation learning from data → Offline RL algorithms
Adaptive Prompting: Ad-hoc Prompt Composition for Social Bias DetectionLLM reasoning → Prompting and chain-of-thought reasoning
WyckoffDiff -- A Generative Diffusion Model for Crystal SymmetryDiffusion models → Flow matching, time series and science
UniMoD: Efficient Unified Multimodal Transformers with Mixture-of-DepthsEfficient LLM inference and serving → Visual token pruning in multimodal models
KARMA: Leveraging Multi-Agent LLMs for Automated Knowledge Graph EnrichmentLLM agents and tool use → Web, search, and research agents
MATH-Perturb: Benchmarking LLMs' Math Reasoning Abilities against Hard PerturbationsLLM reasoning → Mathematical reasoning
Benchmarking Vision-Language Models on Optical Character Recognition in Dynamic Video EnvironmentsMultimodal LLMs / vision-language models → Document understanding and multimodal RAG
FEMBA: Efficient and Scalable EEG Analysis with a Bidirectional Mamba Foundation ModelState-space sequence models (S4, Mamba) → Physiological signal modeling
Prompt-SID: Learning Structural Representation Prompt via Latent Diffusion for Single-Image DenoisingDiffusion models → Inverse problems and image restoration
CoS: Chain-of-Shot Prompting for Long Video UnderstandingMultimodal LLMs / vision-language models → Video understanding with MLLMs
Hybrid State-Space and GRU-based Graph Tokenization Mamba for Hyperspectral Image ClassificationState-space sequence models (S4, Mamba) → Hyperspectral image analysis with Mamba
Occ-LLM: Enhancing Autonomous Driving with Occupancy-Based Large Language ModelsEnd-to-end autonomous driving → Large language models for driving
Habitizing Diffusion Planning for Efficient and Effective Decision MakingDiffusion models → Diffusion and flow policies for robotics +1
AppVLM: A Lightweight Vision Language Model for Online App ControlMultimodal LLMs / vision-language models → Embodied and 3D vision-language models
When Data Manipulation Meets Attack Goals: An In-depth Survey of Attacks for VLMsMultimodal LLMs / vision-language models → MLLM jailbreaks and adversarial robustness
How Humans Help LLMs: Assessing and Incentivizing Human Preference AnnotatorsLLM alignment / RLHF / preference optimisation → Human values and AI feedback alignment
Solving Linear-Gaussian Bayesian Inverse Problems with Decoupled Diffusion Sequential Monte CarloDiffusion models → Inverse problems and image restoration
Hyperparameters in Score-Based Membership Inference AttacksLLM safety and security → Membership inference attacks
Prompt-Tuning Bandits: Enabling Few-Shot Generalization for Efficient Multi-Task Offline RLOffline RL and imitation learning from data → Decision transformers and sequence modeling
Guidance-base Diffusion Models for Improving Photoacoustic Image QualityDiffusion models → Medical imaging and segmentation
LANTERN++: Enhancing Relaxed Speculative Decoding with Static Tree Drafting for Visual Auto-regressive ModelsEfficient LLM inference and serving → Fast autoregressive visual generation
Zero-shot Depth Completion via Test-time Alignment with Affine-invariant Depth PriorDiffusion models → 3D, point cloud and scene generation
Utilizing Novelty-based Evolution Strategies to Train Transformers in Reinforcement LearningOffline RL and imitation learning from data → Decision transformers and sequence modeling +1
Jakiro: Boosting Speculative Decoding with Decoupled Multi-Head via MoELLM reasoning → Test-time compute and efficient thinking +2
DebateBench: A Challenging Long Context Reasoning Benchmark For Large Language ModelsLLM reasoning → Reasoning for agents, planning and code
Emergent Response Planning in LLMsLLM reasoning → Prompting and chain-of-thought reasoning
Confidence Improves Self-Consistency in LLMsLLM reasoning → Test-time compute and efficient thinking
Examining False Positives under Inference Scaling for Mathematical ReasoningLLM reasoning → Mathematical reasoning
C-3PO: Compact Plug-and-Play Proxy Optimization to Achieve Human-like Retrieval-Augmented GenerationRetrieval-augmented generation → Agentic RAG, agents and memory
Right Time to Learn:Promoting Generalization via Bio-inspired Spacing Effect in Knowledge DistillationKnowledge distillation → General knowledge distillation methods
Efficient-vDiT: Efficient Video Diffusion Transformers With Attention TileDiffusion models → Video generation
Optimizing Knowledge Integration in Retrieval-Augmented Generation with Self-SelectionRetrieval-augmented generation → RAG robustness and adaptive retrieval
Animate Anyone 2: High-Fidelity Character Image Animation with Environment AffordanceDiffusion models → Video generation
Self-Correcting Decoding with Generative Feedback for Mitigating Hallucinations in Large Vision-Language ModelsMultimodal LLMs / vision-language models → Hallucination mitigation in LVLMs
Enhanced 3D Object Detection via Diverse Feature Representations of 4D Radar TensorKnowledge distillation → Detection and dense prediction distillation
CSR-Bench: Benchmarking LLM Agents in Deployment of Computer Science Research RepositoriesLLM agents and tool use → Web, search, and research agents
CDM: Contact Diffusion Model for Multi-Contact Point LocalizationDiffusion models → Diffusion and flow policies for robotics
Fine-Tuning is Subgraph Search: A New Lens on Learning DynamicsMechanistic interpretability → Circuits and mechanistic analysis
RALLRec: Improving Retrieval Augmented Large Language Model Recommendation with Representation LearningRetrieval-augmented generation → RAG for recommendation
Fair-MoE: Fairness-Oriented Mixture of Experts in Vision-Language ModelsMultimodal LLMs / vision-language models → Medical vision-language models +1
Debiasing Guidance for Discrete Diffusion with Sequential Monte CarloDiffusion models → Diffusion foundations and sampling
A Planning Framework for Adaptive LabelingDeep reinforcement learning methods → RL theory and sample complexity
Completeness of Datasets Documentation on ML/AI repositories: an Empirical Investigation
What I cannot execute, I do not understand: Training and Evaluating LLMs on Program Execution Traces
Agentic AI and the Cyber Arms Race
Chat-GPT: An AI Based Educational Revolution
Electricity Demand Forecasting in Future Grid States: A Digital Twin-Based Simulation Study
AI Enhanced Ontology Driven NLP for Intelligent Cloud Resource Query Processing Using Knowledge Graphs
SincPD: An Explainable Method based on Sinc Filters to Diagnose Parkinson's Disease Severity by Gait Cycle Analysis
The Case for Cleaner Biosignals: High-fidelity Neural Compressor Enables Transfer from Cleaner iEEG to Noisier EEG

All 410