Papers

New on arXiv

Mon 11 Aug 2025

387 papers

Signature vs. Substance: Evaluating the Balance of Adversarial Resistance and Linguistic Quality in Watermarking Large Language ModelsLLM safety and security → Watermarking generated content and models
Retrieval-Augmented Generation in Industry: An Interview Study on Use Cases, Requirements, Challenges, and EvaluationRetrieval-augmented generation → RAG pipeline design and evaluation
VSF: Simple, Efficient, and Effective Negative Guidance in Few-Step Image Generation Models By Value Sign FlipDiffusion models → Diffusion foundations and sampling
FIDELIS: Blockchain-Enabled Protection Against Poisoning Attacks in Federated LearningLLM safety and security → Federated unlearning and poisoning
Towards Effective MLLM Jailbreaking Through Balanced On-Topicness and OOD-IntensityMultimodal LLMs / vision-language models → MLLM jailbreaks and adversarial robustness +1
MME-Emotion: A Holistic Evaluation Benchmark for Emotional Intelligence in Multimodal Large Language ModelsMultimodal LLMs / vision-language models → Multimodal emotion understanding
SynLLM: A Comparative Analysis of Large Language Models for Medical Tabular Synthetic Data Generation via Prompt EngineeringSynthetic data and self-improvement for LLMs → Medical synthetic data and agents
VISOR: Visual Input-based Steering for Output Redirection in Vision-Language ModelsMultimodal LLMs / vision-language models → Embodied and 3D vision-language models +1
Steerable Pluralism: Pluralistic Alignment via Few-Shot Comparative RegressionLLM alignment / RLHF / preference optimisation → Human values and AI feedback alignment
Re:Verse -- Can Your VLM Read a Manga?Multimodal LLMs / vision-language models → MLLM training and visual instruction tuning +1
GVGAI-LLM: Evaluating Large Language Model Agents with Infinite GamesLLM agents and tool use → Reinforcement learning for LLM agents
POMO+: Leveraging starting nodes in POMO for solving Capacitated Vehicle Routing ProblemDeep reinforcement learning methods → RL for games and combinatorial optimization
MuGa-VTON: Multi-Garment Virtual Try-On via Diffusion Transformers with Prompt CustomizationDiffusion models → 3D, point cloud and scene generation
Beyond Ordinal Preferences: Why Alignment Needs Cardinal Human FeedbackLLM alignment / RLHF / preference optimisation → LLM preference optimization methods
Enhancing Small LLM Alignment through Margin-Based Objective Modifications under Resource ConstraintsLLM alignment / RLHF / preference optimisation → LLM preference optimization methods
Discrete Diffusion-Based Model-Level Explanation of Heterogeneous GNNs with Node FeaturesDiffusion models → Flow matching, time series and science
Architecting Long-Context LLM Acceleration with Packing-Prefetch Scheduler and Ultra-Large Capacity On-Chip MemoriesEfficient LLM inference and serving → Serving systems and request scheduling
Selective KV-Cache Sharing to Mitigate Timing Side-Channels in LLM InferenceEfficient LLM inference and serving → KV cache optimization
Neural Tangent Knowledge Distillation for Optical Convolutional NetworksKnowledge distillation → General knowledge distillation methods
Generating Query-Relevant Document Summaries via Reinforcement LearningDeep reinforcement learning methods → RL training of language and agent models
Mol-R1: Towards Explicit Long-CoT Reasoning in Molecule DiscoveryLLM reasoning → Clinical and biomedical applications
CoDAE: Adapting Large Language Models for Education via Chain-of-Thought Data AugmentationLLM reasoning → Reasoning benchmarks and evaluation
Spatiotemporally Consistent Indoor Lighting Estimation with Diffusion PriorsDiffusion models → 3D, point cloud and scene generation
Scaled-Dot-Product Attention as One-Sided Entropic Optimal TransportDeep reinforcement learning methods → Deep RL algorithms and representations
What Breaks Knowledge Graph based RAG? Benchmarking and Empirical Insights into Reasoning under Incomplete KnowledgeLLM reasoning → Knowledge- and tool-augmented reasoning +1
StableAvatar: Infinite-Length Audio-Driven Avatar Video GenerationDiffusion models → Audio, speech and music
Jinx: Unlimited LLMs for Probing Alignment FailuresLLM alignment / RLHF / preference optimisation → Safety alignment and jailbreak attacks +1
BeyondMimic: From Motion Tracking to Versatile Humanoid Control via Guided DiffusionDiffusion models → Diffusion and flow policies for robotics
Exploring Safety Alignment Evaluation of LLMs in Chinese Mental Health Dialogues via LLM-as-JudgeLLM alignment / RLHF / preference optimisation → Medical LLM alignment
Mamba-FCS: Joint Spatio- Frequency Feature Fusion, Change-Guided Attention, and SeK Loss for Enhanced Semantic Change Detection in Remote SensingState-space sequence models (S4, Mamba) → Point clouds, detection and remote sensing
OMGSR: You Only Need One Mid-timestep Guidance for Real-World Image Super-ResolutionDiffusion models → Inverse problems and image restoration
Capabilities of GPT-5 on Multimodal Medical ReasoningLLM reasoning → Clinical and biomedical applications
Part I: Tricks or Traps? A Deep Dive into RL for LLM ReasoningLLM reasoning → Reinforcement learning for reasoning
SAEMark: Steering Personalized Multilingual LLM Watermarks with Sparse AutoencodersLLM safety and security → Watermarking generated content and models +1
Efficient Speculative Decoding for Llama at Scale: Challenges and SolutionsEfficient LLM inference and serving → Speculative decoding
Reinforcement Learning for Large Model: A SurveyDeep reinforcement learning methods → RL for robot locomotion and manipulation
MedReasoner: Reinforcement Learning Drives Reasoning Grounding from Clinical Thought to Pixel-Level PrecisionLLM reasoning → Clinical and biomedical applications
ReconDreamer-RL: Enhancing Reinforcement Learning via Diffusion-based Scene ReconstructionDiffusion models → Diffusion and flow policies for robotics +2
REX-RAG: Reasoning Exploration with Policy Correction in Retrieval-Augmented GenerationLLM reasoning → Knowledge- and tool-augmented reasoning +1
An effective potential for generative modelling with active matterDiffusion models → Diffusion foundations and sampling
Can LLMs Detect Their Confabulations? Estimating Reliability in Uncertainty-Aware Language ModelsLLM agents and tool use → General LLM agent frameworks and benchmarks
MuaLLM: A Multimodal Large Language Model Agent for Circuit Design Assistance with Hybrid Contextual Retrieval-Augmented GenerationMultimodal LLMs / vision-language models → Document understanding and multimodal RAG +2
Deep Reinforcement Learning with Local Interpretability for Transparent Microgrid Resilience Energy ManagementDeep reinforcement learning methods → RL for energy, networks and operations
BlindGuard: Safeguarding LLM-based Multi-Agent Systems under Unknown AttacksLLM agents and tool use → Agent threats, guardrails and governance
TeamMedAgents: Pareto-Efficient Multi-Agent Medical Reasoning Through Teamwork TheoryLLM reasoning → Clinical and biomedical applications +1
Learned Regularization for Microwave TomographyDiffusion models → Inverse problems and image restoration
ChatGPT on the Road: Leveraging Large Language Model-Powered In-vehicle Conversational Agents for Safer and More Enjoyable Driving ExperienceLLM agents and tool use → Multimodal and embodied agents
TBAC-UniImage: Unified Understanding and Generation by Ladder-Side Diffusion TuningDiffusion models → Image generation, text-to-image and editing
HierSearch: A Hierarchical Enterprise Deep Search Framework Integrating Local and Web SearchesLLM reasoning → Knowledge- and tool-augmented reasoning
ExpVG: Investigating the Design Space of Visual Grounding in Multimodal Large Language ModelMultimodal LLMs / vision-language models → VLM benchmarks and visual reasoning
AdaptFlow: Adaptive Workflow Optimization via Meta-LearningLLM agents and tool use → Software engineering coding agents
Multi-modal Adaptive Mixture of Experts for Cold-start RecommendationMixture-of-experts → MoE for recommender systems
BadPromptFL: A Novel Backdoor Threat to Prompt-based Federated Learning in Multimodal ModelsMultimodal LLMs / vision-language models → Prompt learning and adaptation of CLIP +1
Audio-Thinker: Guiding Audio Language Model When and How to Think via Reinforcement LearningLLM reasoning → Audio-language model reasoning +1
WideSearch: Benchmarking Agentic Broad Info-SeekingLLM agents and tool use → Web, search, and research agents
DIVER: A Multi-Stage Approach for Reasoning-intensive Information RetrievalLLM reasoning → Knowledge- and tool-augmented reasoning +1
The Escalator Problem: Identifying Implicit Motion Blindness in AI for AccessibilityMultimodal LLMs / vision-language models → Video understanding with MLLMs
Beyond Ten Turns: Unlocking Long-Horizon Agentic Search with Large-Scale Asynchronous RLLLM agents and tool use → Web, search, and research agents
WeChat-YATT: A Scalable, Simple, Efficient, and Production Ready Training LibraryLLM alignment / RLHF / preference optimisation → Reward modeling and RLHF
Careful Queries, Credible Results: Teaching RAG Models Advanced Web Search Tools with Reinforcement LearningRetrieval-augmented generation → Agentic RAG, agents and memory
FEAT: A Multi-Agent Forensic AI System with Domain-Adapted Large Language Model for Automated Cause-of-Death AnalysisLLM agents and tool use → Medical and scientific domain agents
Deep Reinforcement Learning with anticipatory reward in LSTM for Collision Avoidance of Mobile RobotsDeep reinforcement learning methods → RL for robot navigation and UAV control
SHIELDA: Structured Handling of Exceptions in LLM-Driven Agentic WorkflowsLLM agents and tool use → Software engineering coding agents
Score Augmentation for Diffusion ModelsDiffusion models → Diffusion foundations and sampling
RSVLM-QA: A Benchmark Dataset for Remote Sensing Vision Language Model-based Question AnsweringMultimodal LLMs / vision-language models → Embodied and 3D vision-language models
MolmoAct: Action Reasoning Models that can Reason in SpaceLLM reasoning → Spatial and embodied reasoning
Diffusing the Blind Spot: Uterine MRI Synthesis with Diffusion ModelsDiffusion models → Medical imaging and segmentation
Multi-agent systems for chemical engineering: A review and perspectiveLLM agents and tool use → Medical and scientific domain agents
EFU: Enforcing Federated Unlearning via Functional EncryptionLLM safety and security → Federated unlearning and poisoning
CATP: Contextually Adaptive Token Pruning for Efficient and Enhanced Multimodal In-Context LearningMultimodal LLMs / vision-language models → Visual token pruning and efficiency +1
CBDES MoE: Hierarchically Decoupled Mixture-of-Experts for Functional Modules in Autonomous DrivingEnd-to-end autonomous driving → VLMs for autonomous driving +1
Effortless Vision-Language Model Specialization in Histopathology without AnnotationMultimodal LLMs / vision-language models → Medical vision-language models
MIMIC: Multimodal Inversion for Model Interpretation and ConceptualizationMultimodal LLMs / vision-language models → MLLM training and visual instruction tuning
Evaluating Large Language Models as Expert AnnotatorsLLM reasoning → Reasoning benchmarks and evaluation
ACD-CLIP: Decoupling Representation and Dynamic Fusion for Zero-Shot Anomaly DetectionMultimodal LLMs / vision-language models → Chart understanding and anomaly detection
DiTVR: Zero-Shot Diffusion Transformer for Video RestorationDiffusion models → Inverse problems and image restoration
EvoCoT: Overcoming the Exploration Bottleneck in Reinforcement LearningLLM reasoning → Reinforcement learning for reasoning +1
Grove MoE: Towards Efficient and Superior MoE LLMs with Adjugate ExpertsMixture-of-experts → MoE large language models
Pareto Multi-Objective Alignment for Language ModelsLLM alignment / RLHF / preference optimisation → LLM preference optimization methods
UniSVG: A Unified Dataset for Vector Graphic Understanding and Generation with Multimodal Large Language ModelsMultimodal LLMs / vision-language models → Chart understanding and anomaly detection
Learning to Align, Aligning to Learn: A Unified Approach for Self-Optimized AlignmentLLM alignment / RLHF / preference optimisation → LLM preference optimization methods
Grouped Speculative Decoding for Autoregressive Image GenerationDiffusion models → Image generation, text-to-image and editing +1
A Tutorial: An Intuitive Explanation of Offline Reinforcement Learning TheoryOffline RL and imitation learning from data → Offline RL algorithms
Chimera: Harnessing Multi-Agent LLMs for Automatic Insider Threat SimulationLLM agents and tool use → Agent threats, guardrails and governance
Separation and Collaboration: Two-Level Routing Grouped Mixture-of-Experts for Multi-Domain Continual LearningMixture-of-experts → Mixture of LoRA experts
SimViews: An Interactive Multi-Agent System Simulating Visitor-to-Visitor Conversational Patterns to Present Diverse Perspectives of Artifacts in Virtual MuseumsLLM agents and tool use → Social simulation and multi-agent debate
Robust Remote Reinforcement Learning over Unreliable Communication Channels using Homomorphic State EncodingDeep reinforcement learning methods → RL theory and sample complexity
Make Your MoVe: Make Your 3D Contents by Adapting Multi-View Diffusion Models to External EditingDiffusion models → 3D, point cloud and scene generation
LoSemB: Logic-Guided Semantic Bridging for Inductive Tool RetrievalLLM agents and tool use → Tool learning and function calling
Risk Map As Middleware: Towards Interpretable Cooperative End-to-end Autonomous Driving for Risk-Aware PlanningEnd-to-end autonomous driving → End-to-end system architecture
TAR: Temporal Anchor-Constrained Reasoning for Video Temporal GroundingLLM reasoning → Multimodal reasoning
Semantic Caching for Low-Cost LLM Serving: From Offline Learning to Online AdaptationEfficient LLM inference and serving → KV cache optimization
1-2-3 Check: Enhancing Contextual Privacy in LLM via Multi-Agent ReasoningLLM reasoning → Reasoning for agents, planning and code +1
Through Their Eyes: User Perceptions on Sensitive Attribute Inference of Social Media Videos by Visual Language ModelsMultimodal LLMs / vision-language models → MLLM jailbreaks and adversarial robustness
GraphCoT-VLA: A 3D Spatial-Aware Reasoning Vision-Language-Action Model for Robotic Manipulation with Ambiguous InstructionsVision-language-action models → 3D and spatial-aware VLA
LaRender: Training-Free Occlusion Control in Image Generation via Latent RenderingDiffusion models → 3D, point cloud and scene generation
Multi-Turn Jailbreaks Are Simpler Than They SeemLLM safety and security → Jailbreaking
Breaking Down and Building Up: Mixture of Skill-Based Vision-and-Language Navigation AgentsLLM agents and tool use → Multimodal and embodied agents
Data Selection for LLM Alignment Using Fine-Grained PreferencesLLM alignment / RLHF / preference optimisation → LLM preference optimization methods
Efficient Approximate Posterior Sampling with Annealed Langevin Monte CarloDiffusion models → Diffusion foundations and sampling

All 387