Papers

New on arXiv

Wed 18 Jun 2025

324 papers

PathCoT: Chain-of-Thought Prompting for Zero-shot Pathology Visual ReasoningLLM reasoning → Clinical and biomedical applications +1
Modulated Diffusion: Accelerating Generative Modeling with Modulated QuantizationDiffusion models → Diffusion foundations and sampling
Evaluating Multimodal Large Language Models on Educational Textbook Question AnsweringMultimodal LLMs / vision-language models → MLLM training and visual instruction tuning
Gazal-R1: Achieving State-of-the-Art Medical Reasoning with Parameter-Efficient Two-Stage TrainingLLM reasoning → Clinical and biomedical applications
PentaRAG: Large-Scale Intelligent Knowledge Retrieval for Enterprise LLM ApplicationsRetrieval-augmented generation → Generation and context integration
FinEval-KR: A Financial Domain Evaluation Framework for Large Language Models' Knowledge and ReasoningLLM reasoning → Reasoning for agents, planning and code
Representation Consistency for Accurate and Coherent LLM Answer AggregationLLM reasoning → Test-time compute and efficient thinking
Poly-Guard: Massive Multi-Domain Safety Policy-Grounded Guardrail DatasetLLM safety and security → Safety alignment, refusal and guardrails
Origins of Creativity in Attention-Based Diffusion ModelsDiffusion models → Diffusion foundations and sampling
MAARTA:Multi-Agentic Adaptive Radiology Teaching AssistantLLM agents and tool use → Medical and scientific domain agents
Context manipulation attacks : Web agents are susceptible to corrupted memoryLLM agents and tool use → Agent threats, guardrails and governance
A Nested Watermark for Large Language ModelsLLM safety and security → Watermarking generated content and models
From RAG to Agentic: Validating Islamic-Medicine Responses with LLM AgentsLLM agents and tool use → Medical and scientific domain agents
Language Models can perform Single-Utterance Self-Correction of Perturbed ReasoningLLM reasoning → Test-time compute and efficient thinking
Fractional Reasoning via Latent Steering Vectors Improves Inference Time ComputeLLM reasoning → Test-time compute and efficient thinking +1
Deep Reinforcement Learning Xiangqi Player with Monte Carlo Tree SearchDeep reinforcement learning methods → RL for games and combinatorial optimization
Visual symbolic mechanisms: Emergent symbol processing in vision language modelsMultimodal LLMs / vision-language models → VLM benchmarks and visual reasoning
CooperRisk: A Driving Risk Quantification Pipeline with Multi-Agent Cooperative Perception and PredictionEnd-to-end autonomous driving → Cooperative and V2X driving
Improving Robotic Manipulation: Techniques for Object Pose Estimation, Accommodating Positional Uncertainty, and Disassembly Tasks from ExamplesDeep reinforcement learning methods → RL for robot locomotion and manipulation
Improving Rectified Flow with Boundary ConditionsDiffusion models → Flow matching, time series and science
MoR: Better Handling Diverse Queries with a Mixture of Sparse, Dense, and Human RetrieversRetrieval-augmented generation → RAG pipeline design and evaluation
Privacy-Preserving in Connected and Autonomous Vehicles Through Vision to Text TransformationEnd-to-end autonomous driving → VLMs for autonomous driving
MEM1: Learning to Synergize Memory and Reasoning for Efficient Long-Horizon AgentsLLM reasoning → Reasoning for agents, planning and code +1
Rethinking LLM Training through Information Geometry and Quantum MetricsLLM reasoning → Reinforcement learning for reasoning
Heterogeneous Federated Reinforcement Learning Using Wasserstein BarycentersDeep reinforcement learning methods → RL for energy, networks and operations
Steering Your Diffusion Policy with Latent Space Reinforcement LearningDiffusion models → Diffusion and flow policies for robotics +1
SLR: Automated Synthesis for Scalable Logical ReasoningLLM reasoning → Reasoning benchmarks and evaluation
VLMInferSlow: Evaluating the Efficiency Robustness of Large Vision-Language Models as a ServiceMultimodal LLMs / vision-language models → MLLM jailbreaks and adversarial robustness
QPPG: Quantum-Preconditioned Policy Gradient for Link Adaptation in Rayleigh Fading ChannelsDeep reinforcement learning methods → Quantum reinforcement learning
Sysformer: Safeguarding Frozen Large Language Models with Adaptive System PromptsLLM safety and security → Safety alignment, refusal and guardrails
D2Diff : A Dual Domain Diffusion Model for Accurate Multi-Contrast MRI SynthesisDiffusion models → Medical imaging and segmentation
InfiniPot-V: Memory-Constrained KV Cache Compression for Streaming Video UnderstandingEfficient LLM inference and serving → KV cache optimization
Nabla-R2D3: Effective and Efficient 3D Diffusion Alignment with 2D RewardsDiffusion models → Diffusion and flow policies for robotics +1
Evolutionary Caching to Accelerate Your Off-the-Shelf Diffusion ModelDiffusion models → Diffusion foundations and sampling +1
GenRecal: Generation after Recalibration from Large to Small Vision-Language ModelsMultimodal LLMs / vision-language models → VLM benchmarks and visual reasoning
Dense SAE Latents Are Features, Not BugsMechanistic interpretability → SAE features and their applications
Embodied Web Agents: Bridging Physical-Digital Realms for Integrated Agent IntelligenceLLM agents and tool use → Multimodal and embodied agents
Leaky Thoughts: Large Reasoning Models Are Not Private ThinkersLLM reasoning → Reasoning for agents, planning and code
SwarmAgentic: Towards Fully Automated Agentic System Generation via Swarm IntelligenceLLM agents and tool use → General LLM agent frameworks and benchmarks
PhishDebate: An LLM-Based Multi-Agent Framework for Phishing Website DetectionLLM agents and tool use → Agent threats, guardrails and governance
cAST: Enhancing Code Retrieval-Augmented Generation with Structural Chunking via Abstract Syntax TreeRetrieval-augmented generation → RAG for code
CAWR: Corruption-Averse Advantage-Weighted Regression for Robust Policy OptimizationOffline RL and imitation learning from data → Offline RL algorithms +1
AutoRule: Reasoning Chain-of-thought Extracted Rule-based Rewards Improve Preference LearningLLM reasoning → Reinforcement learning for reasoning +1
Dual-Stage Value-Guided Inference with Margin-Based Reward Adjustment for Fast and Faithful VLM CaptioningMultimodal LLMs / vision-language models → VLM benchmarks and visual reasoning
Exploring and Exploiting the Inherent Efficiency within Large Reasoning Models for Self-Guided Efficiency EnhancementLLM reasoning → Test-time compute and efficient thinking
Demystifying the Visual Quality Paradox in Multimodal Large Language ModelsMultimodal LLMs / vision-language models → MLLM training and visual instruction tuning
FindingDory: A Benchmark to Evaluate Memory in Embodied AgentsMultimodal LLMs / vision-language models → Embodied and 3D vision-language models +1
Revisiting Compositional Generalization Capability of Large Language Models Considering Instruction Following AbilityLLM reasoning → Reasoning benchmarks and evaluation
The Effect of State Representation on LLM Agent Behavior in Dynamic Routing GamesLLM agents and tool use → Reinforcement learning for LLM agents
LoX: Low-Rank Extrapolation Robustifies LLM Safety Against Fine-tuningLLM safety and security → Safety alignment, refusal and guardrails
One-Step Diffusion for Detail-Rich and Temporally Consistent Video Super-ResolutionDiffusion models → Inverse problems and image restoration
DiscoSG: Towards Discourse-Level Text Scene Graph Parsing through Iterative Graph RefinementMultimodal LLMs / vision-language models → VLM benchmarks and visual reasoning
Intelligent Assistants for the Semiconductor Failure Analysis with LLM-Based Planning AgentsLLM agents and tool use → Software engineering coding agents
Task-Agnostic Experts Composition for Continual LearningLLM reasoning → Reasoning benchmarks and evaluation
Show-o2: Improved Native Unified Multimodal ModelsDiffusion models → Image generation, text-to-image and editing
Stable Gradients for Stable Learning at Scale in Deep Reinforcement LearningDeep reinforcement learning methods → Deep RL algorithms and representations
Diff-TONE: Timestep Optimization for iNstrument Editing in Text-to-Music Diffusion ModelsDiffusion models → Audio, speech and music
RePCS: Diagnosing Data Memorization in LLM-Powered Retrieval-Augmented GenerationRetrieval-augmented generation → Generation and context integration
SPARE: Single-Pass Annotation with Reference-Guided Evaluation for Automatic Process Supervision and Reward ModellingLLM reasoning → Reinforcement learning for reasoning +1
GenHOI: Generalizing Text-driven 4D Human-Object Interaction Synthesis for Unseen ObjectsDiffusion models → Video generation
Multimodal Large Language Models for Medical Report Generation via Customized Prompt TuningMultimodal LLMs / vision-language models → Medical vision-language models
RE-IMAGINE: Symbolic Benchmark Synthesis for Reasoning EvaluationLLM reasoning → Test-time compute and efficient thinking
AgentGroupChat-V2: Divide-and-Conquer Is What LLM-Based Multi-Agent System NeedLLM agents and tool use → General LLM agent frameworks and benchmarks
Zero-Shot Reinforcement Learning Under Partial ObservabilityDeep reinforcement learning methods → Deep RL algorithms and representations
Understanding GUI Agent Localization Biases through Logit SharpnessMultimodal LLMs / vision-language models → Embodied and 3D vision-language models +1
Reward Models in Deep Reinforcement Learning: A SurveyLLM alignment / RLHF / preference optimisation → Reward modeling and RLHF
Provable Maximum Entropy Manifold Exploration via Diffusion ModelsDiffusion models → Diffusion foundations and sampling
When Model Knowledge meets Diffusion Model: Diffusion-assisted Data-free Image Synthesis with Alignment of Domain and ClassDiffusion models → Diffusion foundations and sampling
Efficient Navigation Among Movable Obstacles using a Mobile Manipulator via Hierarchical Policy LearningDeep reinforcement learning methods → RL for robot locomotion and manipulation
Sampling 3D Molecular Conformers with Diffusion TransformersDiffusion models → Flow matching, time series and science
Enhancing One-run Privacy Auditing with Quantile Regression-Based Membership InferenceLLM safety and security → Membership inference attacks
Acoustic Waveform Inversion with Image-to-Image Schr"odinger BridgesDiffusion models → Flow matching, time series and science
Knowledge Distillation Framework for Accelerating High-Accuracy Neural Network-Based Molecular Dynamics SimulationsKnowledge distillation → Graph neural network distillation
One-shot Face Sketch Synthesis in the Wild via Generative Diffusion Prior and Instruction TuningDiffusion models → Image generation, text-to-image and editing
MEGC2025: Micro-Expression Grand Challenge on Spot Then Recognize and Visual Question AnsweringMultimodal LLMs / vision-language models → Multimodal emotion understanding
Human Motion Capture from Loose and Sparse Inertial Sensors with Garment-aware Diffusion ModelsDiffusion models → 3D, point cloud and scene generation
Unlocking Post-hoc Dataset Inference with Synthetic DataSynthetic data and self-improvement for LLMs → LLM synthetic data generation
RAS-Eval: A Comprehensive Benchmark for Security Evaluation of LLM Agents in Real-World EnvironmentsLLM agents and tool use → Agent threats, guardrails and governance
TopClustRAG at SIGIR 2025 LiveRAG ChallengeRetrieval-augmented generation → RAG robustness and adaptive retrieval
Research on Graph-Retrieval Augmented Generation Based on Historical Text Knowledge GraphsRetrieval-augmented generation → Knowledge-graph and structured-knowledge RAG
ProtoReasoning: Prototypes as the Foundation for Generalizable Reasoning in LLMsLLM reasoning → Reasoning benchmarks and evaluation
Privacy-Shielded Image Compression: Defending Against Exploitation from Vision-Language Pretrained ModelsMultimodal LLMs / vision-language models → MLLM jailbreaks and adversarial robustness
From LLMs to MLLMs to Agents: A Survey of Emerging Paradigms in Jailbreak Attacks and Defenses within LLM EcosystemLLM agents and tool use → Agent threats, guardrails and governance +1
LLM Agent for Hyper-Parameter OptimizationLLM agents and tool use → Multimodal and embodied agents
Echo-DND: A dual noise diffusion model for robust and precise left ventricle segmentation in echocardiographyDiffusion models → Medical imaging and segmentation
Robust Instant Policy: Leveraging Student's t-Regression Model for Robust In-context Imitation Learning of Robot ManipulationOffline RL and imitation learning from data → Language-conditioned policies and VLAs
Emergence of Primacy and Recency Effect in Mamba: A Mechanistic Point of ViewState-space sequence models (S4, Mamba) → Core SSM architectures for language
eLLM: Elastic Memory Management Framework for Efficient LLM ServingEfficient LLM inference and serving → KV cache optimization
Booster Gym: An End-to-End Reinforcement Learning Framework for Humanoid Robot LocomotionDeep reinforcement learning methods → RL for robot locomotion and manipulation
CKD-EHR:Clinical Knowledge Distillation for Electronic Health RecordsKnowledge distillation → Knowledge distillation for LLMs
Towards Reliable Forgetting: A Survey on Machine Unlearning VerificationLLM safety and security → Machine unlearning
PDLRecover: Privacy-preserving Decentralized Model Recovery with Machine UnlearningLLM safety and security → Machine unlearning
DyNaVLM: Zero-Shot Vision-Language Navigation System with Dynamic Viewpoints and Self-Refining Graph MemoryMultimodal LLMs / vision-language models → Embodied and 3D vision-language models
Make Your AUV Adaptive: An Environment-Aware Reinforcement Learning Framework For Underwater TasksDeep reinforcement learning methods → RL for robot navigation and UAV control
Learning-Time Encoding Shapes Unlearning in LLMsLLM safety and security → Machine unlearning
Semantically-Aware Rewards for Open-Ended R1 Training in Free-Form GenerationLLM alignment / RLHF / preference optimisation → Reward modeling and RLHF
HEAL: An Empirical Study on Hallucinations in Embodied Agents Driven by Large Language ModelsLLM agents and tool use → Medical and scientific domain agents
Sequential Policy Gradient for Adaptive Hyperparameter OptimizationDeep reinforcement learning methods → RL theory and sample complexity
Truncated Proximal Policy OptimizationLLM reasoning → Reinforcement learning for reasoning +1
PCS Workflow for Veridical Data Science in the Age of AI

All 324