Papers

New on arXiv

Wed 27 Aug 2025

330 papers

From Search to Reasoning: A Five-Level RAG Capability Framework for Enterprise DataRetrieval-augmented generation → RAG pipeline design and evaluation
Generalizable Pareto-Optimal Offloading with Reinforcement Learning in Mobile Edge ComputingDeep reinforcement learning methods → RL for energy, networks and operations
Do MLLMs Really Understand the Charts?LLM reasoning → Multimodal reasoning +1
AR2: Adversarial Reinforcement Learning for Abstract Reasoning in Large Language ModelsLLM reasoning → Reinforcement learning for reasoning
PG-Agent: An Agent Powered by Page GraphMultimodal LLMs / vision-language models → Embodied and 3D vision-language models +1
MODE: Mixture of Document Experts for RAGRetrieval-augmented generation → RAG pipeline design and evaluation
Financial Decision Making using Reinforcement Learning with Dirichlet Priors and Quantum-Inspired Genetic OptimizationDeep reinforcement learning methods → Adversarial robustness and RL in finance
Ensemble Debates with Local Large Language Models for AI AlignmentLLM alignment / RLHF / preference optimisation → LLM preference optimization methods
AEGIS : Automated Co-Evolutionary Framework for Guarding Prompt Injections SchemaLLM safety and security → Prompt injection and agent security
Learning to Refine: Self-Refinement of Parallel Reasoning in LLMsLLM reasoning → Test-time compute and efficient thinking
ERTACache: Error Rectification and Timesteps Adjustment for Efficient DiffusionDiffusion models → Diffusion foundations and sampling
Can LLMs Generate Behaviors for Embodied Virtual Agents Based on Personality Traits?LLM agents and tool use → Social simulation and multi-agent debate
Can Compact Language Models Search Like Agents? Distillation-Guided Policy Optimization for Preserving Agentic RAG CapabilitiesLLM agents and tool use → Reinforcement learning for LLM agents
Dynamics-Aligned Latent Imagination in Contextual World Models for Zero-Shot GeneralizationWorld models → Latent dynamics models and model-based RL +1
Network-Level Prompt and Trait Leakage in Local Research AgentsLLM agents and tool use → Agent threats, guardrails and governance
How Multimodal LLMs Solve Image Tasks: A Lens on Visual Grounding, Task Reasoning, and Answer DecodingLLM reasoning → Multimodal reasoning +1
Predictable LLM Serving on GPU ClustersEfficient LLM inference and serving → Serving systems and request scheduling
AI reasoning effort predicts human decision time in content moderationLLM reasoning → Reasoning benchmarks and evaluation
Validating Generative Agent-Based Models for Logistics and Supply Chain Management ResearchLLM agents and tool use → Social simulation and multi-agent debate
ATMS-KD: Adaptive Temperature and Mixed Sample Knowledge Distillation for a Lightweight Residual CNN in Agricultural Embedded SystemsKnowledge distillation → Detection and dense prediction distillation
Robustness Assessment and Enhancement of Text Watermarking for Google's SynthIDLLM safety and security → Watermarking generated content and models
A Novel Framework for Automated Explain Vision Model Using Vision-Language ModelsMultimodal LLMs / vision-language models → VLM benchmarks and visual reasoning
The Role of Teacher Calibration in Knowledge DistillationKnowledge distillation → General knowledge distillation methods
Grounding Multimodal Large Language Models with Quantitative Skin Attributes: A Retrieval StudyMultimodal LLMs / vision-language models → Medical vision-language models
SDiFL: Stable Diffusion-Driven Framework for Image Forgery LocalizationDiffusion models → Image generation, text-to-image and editing
Mitigating Hallucinations in Multimodal LLMs via Object-aware Preference OptimizationMultimodal LLMs / vision-language models → Hallucination mitigation in LVLMs +1
IntentionReasoner: Facilitating Adaptive LLM Safeguards through Intent Reasoning and Selective Query RefinementLLM reasoning → Reasoning for agents, planning and code +1
MicroLad: 2D-to-3D Microstructure Reconstruction and Generation via Latent Diffusion and Score DistillationDiffusion models → Flow matching, time series and science
CODA: Coordinating the Cerebrum and Cerebellum for a Dual-Brain Computer Use Agent with Decoupled Reinforcement LearningLLM agents and tool use → GUI and mobile agents +1
Discrete-Guided Diffusion for Scalable and Safe Multi-Robot Motion PlanningDiffusion models → Diffusion and flow policies for robotics
DisarmRAG: Stealthy Retriever-Centric Poisoning to Disable Self-Correction in Retrieval-Augmented Generation (Extended Version)Retrieval-augmented generation → Privacy and security of RAG
Discrete Diffusion VLA: Bringing Discrete Diffusion to Action Decoding in Vision-Language-Action PoliciesVision-language-action models → Efficient VLA inference and adaptation +2
11Plus-Bench: Demystifying Multimodal LLM Spatial Reasoning with Cognitive-Inspired AnalysisLLM reasoning → Spatial and embodied reasoning +1
Neural Conditional Simulation for Complex Spatial ProcessesDiffusion models → Flow matching, time series and science
Reinforcement Learning for Search Tree Size Minimization in Constraint Programming: New Results on Scheduling BenchmarksDeep reinforcement learning methods → RL for energy, networks and operations
Forewarned is Forearmed: Pre-Synthesizing Jailbreak-like Instructions to Enhance LLM Safety Guardrail to Potential AttacksLLM alignment / RLHF / preference optimisation → Safety alignment and jailbreak attacks +1
Pruning Strategies for Backdoor Defense in LLMsLLM safety and security → Backdoor and data poisoning attacks
GS: Generative Segmentation via Label DiffusionDiffusion models → Medical imaging and segmentation
Symphony: A Decentralized Multi-Agent Framework for Scalable Collective IntelligenceLLM agents and tool use → General LLM agent frameworks and benchmarks
SWIRL: A Staged Workflow for Interleaved Reinforcement Learning in Mobile GUI ControlMultimodal LLMs / vision-language models → Embodied and 3D vision-language models +1
CataractSurg-80K: Knowledge-Driven Benchmarking for Structured Reasoning in Ophthalmic Surgery PlanningLLM reasoning → Clinical and biomedical applications
AgentCoMa: A Compositional Benchmark Mixing Commonsense and Mathematical Reasoning in Real-World ScenariosLLM reasoning → Reasoning benchmarks and evaluation
Diffusion Language Models Know the Answer Before DecodingDiffusion models → Discrete and text diffusion
Evaluating Language Model Reasoning about Confidential InformationLLM reasoning → Reasoning for agents, planning and code
GLSim: Detecting Object Hallucinations in LVLMs via Global-Local SimilarityMultimodal LLMs / vision-language models → Hallucination mitigation in LVLMs
Assessing the Geolocation Capabilities, Limitations and Societal Risks of Generative Vision-Language ModelsMultimodal LLMs / vision-language models → Embodied and 3D vision-language models
Long-VLA: Unleashing Long-Horizon Capability of Vision Language Action Model for Robot ManipulationVision-language-action models → Generalist VLA models and benchmarks
KRETA: A Benchmark for Korean Reading and Reasoning in Text-Rich VQA Attuned to Diverse Visual ContextsLLM reasoning → Multimodal reasoning +1
Socially Interactive Agents for Preserving and Transferring Tacit Knowledge in OrganizationsLLM agents and tool use → Social simulation and multi-agent debate
CASE: An Agentic AI Framework for Enhancing Scam Intelligence in Digital PaymentsLLM agents and tool use → Agent threats, guardrails and governance
HEAL: A Hypothesis-Based Preference-Aware Analysis FrameworkLLM alignment / RLHF / preference optimisation → LLM preference optimization methods
Your AI Bosses Are Still Prejudiced: The Emergence of Stereotypes in LLM-Based Multi-Agent SystemsLLM agents and tool use → Social simulation and multi-agent debate
Ontology-Based Concept Distillation for Radiology Report Retrieval and LabelingRetrieval-augmented generation → Medical and biomedical RAG
Logical Reasoning with Outcome Reward Models for Test-Time ScalingLLM reasoning → Reasoning benchmarks and evaluation +1
Adaptive Scaling of Policy Constraints for Offline Reinforcement LearningOffline RL and imitation learning from data → Offline RL algorithms
The Information Dynamics of Generative DiffusionDiffusion models → Diffusion foundations and sampling
Youtu-GraphRAG: Vertically Unified Agents for Graph Retrieval-Augmented Complex ReasoningLLM reasoning → Knowledge- and tool-augmented reasoning +2
Tracking World States with Language Models: State-Based Evaluation Using ChessWorld models → World models for language agents
Memory-R1: Enhancing Large Language Model Agents to Manage and Utilize Memories via Reinforcement LearningLLM agents and tool use → Agent memory systems
Analysing Chain of Thought Dynamics: Active Guidance or Unfaithful Post-hoc Rationalisation?LLM reasoning → Prompting and chain-of-thought reasoning
T2R-bench: A Benchmark for Generating Article-Level Reports from Real World Industrial TablesLLM reasoning → Knowledge- and tool-augmented reasoning
Bootstrapping Learned Cost Models with Synthetic SQL QueriesSynthetic data and self-improvement for LLMs → Synthetic reasoning data and bootstrapping
StableIntrinsic: Detail-preserving One-step Diffusion Model for Multi-view Material EstimationDiffusion models → 3D, point cloud and scene generation
Fast 3D Diffusion for Scalable Granular Media SynthesisDiffusion models → Flow matching, time series and science
Spotlight Attention: Towards Efficient LLM Generation via Non-linear Hashing-based KV Cache RetrievalEfficient LLM inference and serving → KV cache optimization
Safety Alignment Should Be Made More Than Just A Few Attention HeadsLLM alignment / RLHF / preference optimisation → Safety alignment and jailbreak attacks +1
InquireMobile: Teaching VLM-based Mobile Agent to Request Human Assistance via Reinforcement Fine-TuningMultimodal LLMs / vision-language models → Embodied and 3D vision-language models +1
Self-Rewarding Vision-Language Model via Reasoning DecompositionLLM reasoning → Multimodal reasoning +2
FakeSV-VLM: Taming VLM for Detecting Fake Short-Video News via Progressive Mixture-Of-Experts AdapterMixture-of-experts → Multimodal MoE models
Divide, Weight, and Route: Difficulty-Aware Optimization with Dynamic Expert Fusion for Long-tailed RecognitionMixture-of-experts → Vision mixture-of-experts
LFD: Layer Fused Decoding to Exploit External Knowledge in Retrieval-Augmented GenerationRetrieval-augmented generation → RAG robustness and adaptive retrieval
Instructional Agents: Reducing Teaching Faculty Workload through Multi-Agent Instructional DesignLLM agents and tool use → Educational agents and math problem solving
Impedance Primitive-augmented Hierarchical Reinforcement Learning for Sequential TasksDeep reinforcement learning methods → RL for robot locomotion and manipulation
Encouraging Good Processes Without the Need for Good Answers: Reinforcement Learning for LLM Agent PlanningLLM agents and tool use → Reinforcement learning for LLM agents
Understanding and Leveraging the Expert Specialization of Context Faithfulness in Mixture-of-Experts LLMsMixture-of-experts → Expert routing strategies
Guiding Noisy Label Conditional Diffusion Models with Score-based Discriminator CorrectionDiffusion models → Diffusion foundations and sampling
ReST-RL: Reinforcing LLM Reasoning through Unified Self-Training and Value-Guided SearchLLM reasoning → Reinforcement learning for reasoning +1
Generative Models for Synthetic Data: Transforming Data Mining in the GenAI EraDiffusion models → Diffusion foundations and sampling +1
Counterfactual Reward Model Training for Bias Mitigation in Multimodal Reinforcement LearningLLM alignment / RLHF / preference optimisation → Reward modeling and RLHF
Functional Consistency of LLM Code Embeddings: A Self-Evolving Data Synthesis Framework for BenchmarkingSynthetic data and self-improvement for LLMs → Code and instruction data synthesis
MobText-SISA: Efficient Machine Unlearning for Mobility Logs with Spatio-Temporal and Natural-Language DataLLM safety and security → Machine unlearning
CVBench: Benchmarking Cross-Video Synergies for Complex Multimodal ReasoningLLM reasoning → Multimodal reasoning +1
Alignment with Fill-In-the-Middle for Enhancing Code GenerationLLM alignment / RLHF / preference optimisation → Process reward models for reasoning
Blockwise SFT for Diffusion Language Models: Reconciling Bidirectional Attention and Autoregressive DecodingDiffusion models → Discrete and text diffusion
MotionFlux: Efficient Text-Guided Motion Generation through Rectified Flow Matching and Preference AlignmentDiffusion models → Video generation +1
DATR: Diffusion-based 3D Apple Tree Reconstruction Framework with Sparse-ViewDiffusion models → 3D, point cloud and scene generation
A Self-Supervised Mixture-of-Experts Framework for Multi-behavior RecommendationMixture-of-experts → MoE for recommender systems
Learning Game-Playing Agents with Generative Code OptimizationLLM agents and tool use → Reinforcement learning for LLM agents
Aegis: Taxonomy and Optimizations for Overcoming Agent-Environment Failures in LLM AgentsLLM agents and tool use → General LLM agent frameworks and benchmarks
SLIM: Subtrajectory-Level Elimination for More Effective ReasoningLLM reasoning → Test-time compute and efficient thinking
Servant, Stalker, Predator: How An Honest, Helpful, And Harmless (3H) Agent Unlocks Adversarial SkillsLLM agents and tool use → Agent threats, guardrails and governance
Mind the Third Eye! Benchmarking Privacy Awareness in MLLM-powered Smartphone AgentsLLM agents and tool use → Agent threats, guardrails and governance
Rule Synergy Analysis using LLMs: State of the Art and ImplicationsLLM reasoning → Reasoning for agents, planning and code
LegiScout: A Visual Tool for Understanding Complex Legislation
Tokenization Strategies for Low-Resource Agglutinative Languages in Word2Vec: Case Study on Turkish and Finnish
The LLM as a Network Operator: A Vision for Generative AI in the 6G Radio Access Network
DCHO: A Decomposition-Composition Framework for Predicting Higher-Order Brain Connectivity to Enhance Diverse Downstream Applications
Machine-learning competition to grade EEG background patterns in newborns with hypoxic-ischaemic encephalopathy
Personalized Sleep Prediction via Deep Adaptive Spatiotemporal Modeling and Sparse Data
Uncertainty-Aware Collaborative System of Large and Small Models for Multimodal Sentiment Analysis

All 330