Papers

New on arXiv

Fri 10 Oct 2025

455 papers

Context-Aware Visual Prompting: Automating Geospatial Web Dashboards with Large Language Models and Agent Self-Validation for Decision SupportLLM agents and tool use → Multimodal and embodied agents
Modeling Layered Consciousness with Multi-Agent Large Language ModelsLLM agents and tool use → Social simulation and multi-agent debate
How Good Are LLMs at Processing Tool Outputs?LLM agents and tool use → Tool learning and function calling
ATLAS: Adaptive Trading with LLM AgentS Through Dynamic Prompt Optimization and Multi-Agent CoordinationLLM agents and tool use → Financial trading and forecasting agents
VisuoAlign: Safety Alignment of LVLMs with Multimodal Tree SearchLLM reasoning → Reasoning for agents, planning and code +3
Users as Annotators: LLM Preference Learning from Comparison ModeLLM alignment / RLHF / preference optimisation → Human values and AI feedback alignment
A Linguistics-Aware LLM Watermarking via Syntactic PredictabilityLLM safety and security → Watermarking generated content and models
Bridging the Semantic Gap: Contrastive Rewards for Multilingual Text-to-SQL with GRPOLLM reasoning → Knowledge- and tool-augmented reasoning
AI Agents for the Dhumbal Card Game: A Comparative StudyDeep reinforcement learning methods → RL for games and combinatorial optimization
HeadsUp! High-Fidelity Portrait Image Super-ResolutionDiffusion models → Inverse problems and image restoration
Don't Throw Away Your Pretrained ModelLLM reasoning → Reasoning benchmarks and evaluation
Agentic Property-Based Testing: Finding Bugs Across the Python EcosystemLLM agents and tool use → Software engineering coding agents
The Personalization Trap: How User Memory Alters Emotional Reasoning in LLMsLLM reasoning → Audio-language model reasoning
Autonomous Agents for Scientific Discovery: Orchestrating Scientists, Language, Code, and PhysicsLLM agents and tool use → Web, search, and research agents
Overconfident and Blind to Details: Fixing Prompt Insensitivity with Abductive Preference LearningLLM alignment / RLHF / preference optimisation → LLM preference optimization methods
DELTA: Dynamic Layer-Aware Token Attention for Efficient Long-Context ReasoningLLM reasoning → Test-time compute and efficient thinking
WARC-Bench: Web Archive Based Benchmark for GUI Subtask ExecutionsLLM agents and tool use → Web, search, and research agents
CoBia: Constructed Conversations Can Trigger Otherwise Concealed Societal Biases in LLMsLLM safety and security → Safety alignment, refusal and guardrails
Cluster-Aware Prompt Ensemble Learning for Few-Shot Vision-Language Model AdaptationMultimodal LLMs / vision-language models → Prompt learning and adaptation of CLIP
Text Prompt Injection of Vision Language ModelsMultimodal LLMs / vision-language models → MLLM jailbreaks and adversarial robustness +1
How can we assess human-agent interactions? Case studies in software agent designLLM agents and tool use → Software engineering coding agents
Causality neq Decodability, and Vice Versa: Lessons from Interpreting Counting ViTsMechanistic interpretability → Circuits and mechanistic analysis
Enhancing Diffusion Policy with Classifier-Free Guidance for Temporal Robotic TasksDiffusion models → Diffusion and flow policies for robotics
The Geometry of Reasoning: Flowing Logics in Representation SpaceLLM reasoning → Reasoning benchmarks and evaluation
Building a Foundational Guardrail for General Agentic Systems via Synthetic DataLLM agents and tool use → Agent threats, guardrails and governance +2
GOLD PANNING: Strategic Context Shuffling for Needle-in-Haystack ReasoningLLM reasoning → Knowledge- and tool-augmented reasoning
PatentVision: A multimodal method for drafting patent applicationsMultimodal LLMs / vision-language models → Chart understanding and anomaly detection
Constructive Distortion: Improving MLLMs with Attention-Guided Image WarpingMultimodal LLMs / vision-language models → MLLM training and visual instruction tuning
VisRAG2.0: Mitigating Visual Hallucinations via Evidence-Guided Multi-Image Reasoning in Visual Retrieval-Augmented GenerationLLM reasoning → Multimodal reasoning +2
A Comprehensive Survey on Benchmarks and Solutions in Software Engineering of LLM-Empowered Agentic SystemLLM agents and tool use → Software engineering coding agents
Preference-Aware Memory Update for Long-Term LLM AgentsLLM reasoning → Reasoning for agents, planning and code +1
All Code, No Thought: Current Language Models Struggle to Reason in Ciphered LanguageLLM reasoning → Reasoning for agents, planning and code
SeCon-RAG: A Two-Stage Semantic Filtering and Conflict-Free Framework for Trustworthy RAGRetrieval-augmented generation → RAG robustness and adaptive retrieval
StreamingVLM: Real-Time Understanding for Infinite Video StreamsMultimodal LLMs / vision-language models → Video understanding with MLLMs
VITA-VLA: Efficiently Teaching Vision-Language Models to Act via Action Expert DistillationMultimodal LLMs / vision-language models → Embodied and 3D vision-language models +1
SpaceVista: All-Scale Visual Spatial Reasoning from mm to kmLLM reasoning → Spatial and embodied reasoning
Prompting Test-Time Scaling Is A Strong LLM Reasoning Data AugmentationLLM reasoning → Test-time compute and efficient thinking
Mind-Paced Speaking: A Dual-Brain Approach to Real-Time Reasoning in Spoken Language ModelsLLM reasoning → Audio-language model reasoning
Vision Language Models: A Survey of 26K PapersMultimodal LLMs / vision-language models → VLM benchmarks and visual reasoning
Dyna-Mind: Learning to Simulate from Experience for Better AI AgentsLLM reasoning → Reasoning for agents, planning and code
Safe, Untrusted, "Proof-Carrying" AI Agents: toward the agentic lakehouseLLM agents and tool use → Agent threats, guardrails and governance
TC-LoRA: Temporally Modulated Conditional LoRA for Adaptive Diffusion ControlDiffusion models → Diffusion foundations and sampling
A Comprehensive Evaluation of Multilingual Chain-of-Thought Reasoning: Performance, Consistency, and Faithfulness Across LanguagesLLM reasoning → Reasoning benchmarks and evaluation
Beyond Surface Reasoning: Unveiling the True Long Chain-of-Thought Capacity of Diffusion Large Language ModelsLLM reasoning → Diffusion reasoning and test-time scaling
Guiding Energy-Efficient Locomotion through Impact Mitigation RewardsDeep reinforcement learning methods → RL for robot locomotion and manipulation
SPG: Sandwiched Policy Gradient for Masked Diffusion Language ModelsDiffusion models → Discrete and text diffusion
Mitigating Overthinking through Reasoning ShapingLLM reasoning → Test-time compute and efficient thinking
Conditional Flow Matching for Bayesian Posterior InferenceDiffusion models → Flow matching, time series and science
StatEval: A Comprehensive Benchmark for Large Language Models in StatisticsLLM reasoning → Reasoning benchmarks and evaluation
MRMR: A Realistic and Expert-Level Multidisciplinary Benchmark for Reasoning-Intensive Multimodal RetrievalLLM reasoning → Multimodal reasoning
Toward Autonomous Soft Robotic Endovascular Navigation via Imitation LearningOffline RL and imitation learning from data → Safe offline RL
Near-Optimal Second-Order Guarantees for Model-Based Adversarial Imitation LearningOffline RL and imitation learning from data → Imitation learning methods
FOGMACHINE -- Leveraging Discrete-Event Simulation and Scene Graphs for Modeling Hierarchical, Interconnected Environments under Partial Observations from Mobile AgentsLLM agents and tool use → Multimodal and embodied agents
Multimodal Policy Internalization for Conversational AgentsLLM agents and tool use → General LLM agent frameworks and benchmarks
D-TPT: Dimensional Entropy Maximization for Calibrating Test-Time Prompt Tuning in Vision-Language ModelsMultimodal LLMs / vision-language models → Prompt learning and adaptation of CLIP
Hybrid Models for Natural Language Reasoning: The Case of Syllogistic LogicLLM reasoning → Reasoning benchmarks and evaluation
Getting Your Indices in a Row: Full-Text Search for LLM Training Data for Real WorldLLM safety and security → Machine unlearning
Adaptive Attacks on Trusted Monitors Subvert AI Control ProtocolsLLM agents and tool use → Agent threats, guardrails and governance +1
Failure Prediction at Runtime for Generative Robot PoliciesDiffusion models → Diffusion and flow policies for robotics +1
KORMo: Korean Open Reasoning Model for EveryoneLLM reasoning → Reasoning benchmarks and evaluation
Agentic Systems in Radiology: Design, Applications, Evaluation, and ChallengesLLM agents and tool use → Medical and scientific domain agents
Beyond Single-Granularity Prompts: A Multi-Scale Chain-of-Thought Prompt Learning for GraphLLM reasoning → Prompting and chain-of-thought reasoning
Don't Tell the Answer, Truly Guide the Reasoning During RL RolloutsLLM reasoning → Reinforcement learning for reasoning
Token-Level Policy Optimization: Linking Group-Level Rewards to Token-Level Aggregation via Markov LikelihoodLLM reasoning → Reinforcement learning for reasoning
Minkowski-MambaNet: A Point Cloud Framework with Selective State Space Models for Forest Biomass QuantificationState-space sequence models (S4, Mamba) → Point clouds, detection and remote sensing
BLINK-Twice: You see, but do you observe? A Reasoning Benchmark on Visual PerceptionLLM reasoning → Multimodal reasoning +1
Boosting Multi-modal Keyphrase Prediction with Dynamic Chain-of-Thought in Vision-Language ModelsLLM reasoning → Multimodal reasoning +1
Logit Arithmetic Elicits Long Reasoning Capabilities Without TrainingLLM reasoning → Test-time compute and efficient thinking +1
ReTraceQA: Evaluating Reasoning Traces of Small Language Models in Commonsense Question AnsweringLLM reasoning → Reasoning benchmarks and evaluation
Deductive Logic in Language Models: Horizontal vs Vertical ReasoningLLM reasoning → Reasoning benchmarks and evaluation
Efficient Bayesian Inference from Noisy Pairwise ComparisonsLLM alignment / RLHF / preference optimisation → Human values and AI feedback alignment
FLRC: Fine-grained Low-Rank Compressor for Efficient LLM InferenceEfficient LLM inference and serving → KV cache optimization
Safety Game: Inference-Time Alignment of Black-Box LLMs via Constrained OptimizationLLM alignment / RLHF / preference optimisation → Safety alignment and jailbreak attacks +1
Rate optimal learning of equilibria from dataOffline RL and imitation learning from data → Imitation learning methods
RadioFlow: Efficient Radio Map Construction Framework with Flow MatchingDiffusion models → Diffusion for compression and wireless
Verifying Chain-of-Thought Reasoning via Its Computational GraphLLM reasoning → Prompting and chain-of-thought reasoning
Mask Tokens as Prophet: Fine-Grained Cache Eviction for Efficient dLLM InferenceEfficient LLM inference and serving → Diffusion LLM and MoE inference
CapGeo-Bench: Decoupling Visual Perception from Reasoning and Evaluating Geometric UnderstandingLLM reasoning → Spatial and embodied reasoning +1
Spotlight on Token Perception for Multimodal Reinforcement LearningLLM reasoning → Multimodal reasoning +2
CLARity: Reasoning Consistency Alone Can Teach Reinforced ExpertsLLM reasoning → Reinforcement learning for reasoning
Inflated Excellence or True Performance? Rethinking Medical Diagnostic Benchmarks with Dynamic EvaluationLLM reasoning → Clinical and biomedical applications
MomentSeg: Moment-Centric Sampling for Enhanced Video Pixel UnderstandingLLM reasoning → Multimodal reasoning
Goal-oriented Backdoor Attack against Vision-Language-Action Models via Physical ObjectsVision-language-action models → Safety and adversarial attacks +1
CFVBench: A Comprehensive Video Benchmark for Fine-grained Multimodal Retrieval-Augmented GenerationMultimodal LLMs / vision-language models → Video understanding with MLLMs +1
GREAT: Generalizable Backdoor Attacks in RLHF via Emotion-Aware Trigger SynthesisLLM alignment / RLHF / preference optimisation → Safety alignment and jailbreak attacks
Hallucination Filtering in Radiology Vision-Language Models Using Discrete Semantic EntropyMultimodal LLMs / vision-language models → Hallucination mitigation in LVLMs
DSPO: Stable and Efficient Policy Optimization for Agentic Search and ReasoningLLM reasoning → Knowledge- and tool-augmented reasoning +1
Obstacle Avoidance using Dynamic Movement Primitives and Reinforcement LearningDeep reinforcement learning methods → RL for robot navigation and UAV control
On the Robustness of Vision-Language Models in Zero-shot Privacy ClassificationMultimodal LLMs / vision-language models → MLLM jailbreaks and adversarial robustness
Application of Deep Reinforcement Learning to At-the-Money S&P 500 Options HedgingDeep reinforcement learning methods → Adversarial robustness and RL in finance
Fundamentals of Building Autonomous LLM AgentsLLM agents and tool use → General LLM agent frameworks and benchmarks
RegexPSPACE: A Benchmark for Evaluating LLM Reasoning on PSPACE-complete Regex ProblemsLLM reasoning → Reasoning benchmarks and evaluation
FM-IRL: Flow-Matching for Reward Modeling and Policy Regularization in Reinforcement LearningDiffusion models → Diffusion and flow policies for robotics +3
Stable Video Infinity: Infinite-Length Video Generation with Error RecyclingDiffusion models → Video generation
DICE: Structured Reasoning in LLMs through SLM-Guided Chain-of-Thought CorrectionLLM reasoning → Prompting and chain-of-thought reasoning
Flow-Opt: Scalable Centralized Multi-Robot Trajectory Optimization with Flow Matching and Differentiable OptimizationDiffusion models → Diffusion and flow policies for robotics
LLaMAX2: Your Translation-Enhanced Model also Performs Well in ReasoningLLM reasoning → Reasoning benchmarks and evaluation
Student Development Agent: Risk-free Simulation for Evaluating AIED InnovationsLLM agents and tool use → Educational agents and math problem solving
Augmenting Dialog with Think-Aloud Utterances for Modeling Individual Personality Traits by LLMLLM reasoning → Reasoning benchmarks and evaluation
Agentic-KGR: Co-evolutionary Knowledge Graph Construction through Multi-Agent Reinforcement LearningLLM agents and tool use → Web, search, and research agents

All 455