Papers

New on arXiv

Fri 14 Nov 2025

403 papers

Revisiting Disaggregated Large Language Model Serving for Performance and Energy ImplicationsEfficient LLM inference and serving → Serving systems and request scheduling
Closing the Gap: Data-Centric Fine-Tuning of Vision Language Models for the Standardized Exam QuestionsLLM reasoning → Multimodal reasoning +1
Structured Definitions and Segmentations for Legal Reasoning in LLMs: A Study on Indian Legal DataLLM reasoning → Reasoning for agents, planning and code
PIRA: Preference-Oriented Instruction-Tuned Reward Models with Dual AggregationLLM alignment / RLHF / preference optimisation → Reward modeling and RLHF
Enhancing Robustness of Offline Reinforcement Learning Under Data Corruption via Sharpness-Aware MinimizationOffline RL and imitation learning from data → Offline RL algorithms
Spatial Reasoning in Multimodal Large Language Models: A Survey of Tasks, Benchmarks and MethodsLLM reasoning → Spatial and embodied reasoning +1
Dynamic Temperature Scheduler for Knowledge DistillationKnowledge distillation → General knowledge distillation methods
PROF: An LLM-based Reward Code Preference Optimization Framework for Offline Imitation LearningLLM alignment / RLHF / preference optimisation → Reward modeling and RLHF +1
Knowledge vs. Experience: Asymptotic Limits of Impatience in Edge TenantsDeep reinforcement learning methods → Deep RL algorithms and representations
MoETTA: Test-Time Adaptation Under Mixed Distribution Shifts with MoE-LayerNormMixture-of-experts → Efficient MoE training and inference systems
Multi-Agent VLMs Guided Self-Training with PNU Loss for Low-Resource Offensive Content DetectionLLM agents and tool use → Social simulation and multi-agent debate +1
From Events to Clarity: The Event-Guided Diffusion Framework for DehazingDiffusion models → Inverse problems and image restoration
InData: Towards Secure Multi-Step, Tool-Based Data AnalysisLLM agents and tool use → Web, search, and research agents
MATT-Diff: Multimodal Active Target Tracking by Diffusion PolicyDiffusion models → Diffusion and flow policies for robotics +1
Looking Forward: Challenges and Opportunities in Agentic AI ReliabilityLLM agents and tool use → Agent threats, guardrails and governance
An Analysis of Architectural Impact on LLM-based Abstract Visual Reasoning: A Systematic Benchmark on RAVEN-FAIRLLM reasoning → Multimodal reasoning
Forgetting-MarI: LLM Unlearning via Marginal Information RegularizationLLM safety and security → Machine unlearning
Seeing the Forest and the Trees: Query-Aware Tokenizer for Long-Video Multimodal Language ModelsMultimodal LLMs / vision-language models → Video understanding with MLLMs
KVSwap: Disk-aware KV Cache Offloading for Long-Context On-device InferenceEfficient LLM inference and serving → KV cache optimization
Prompt Triage: Structured Optimization Enhances Vision-Language Model Performance on Medical Imaging BenchmarksMultimodal LLMs / vision-language models → Medical vision-language models
VULPO: Context-Aware Vulnerability Detection via On-Policy LLM OptimizationLLM reasoning → Reasoning for agents, planning and code
Chain-of-Generation: Progressive Latent Diffusion for Text-Guided Molecular DesignDiffusion models → Flow matching, time series and science
Better LLM Reasoning via Dual-PlayLLM reasoning → Reinforcement learning for reasoning
Graph-Based Imitation and Reinforcement Learning for Efficient Benders DecompositionDeep reinforcement learning methods → RL for games and combinatorial optimization
LAVQA: A Latency-Aware Visual Question Answering Framework for Shared Autonomy in Self-Driving VehiclesEnd-to-end autonomous driving → VLMs for autonomous driving
TopoPerception: A Shortcut-Free Evaluation of Global Visual Perception in Large Vision-Language ModelsMultimodal LLMs / vision-language models → VLM benchmarks and visual reasoning
Conformal Constrained Policy Optimization for Cost-Effective LLM AgentsLLM agents and tool use → Reinforcement learning for LLM agents
On the Notion that Language Models ReasonLLM reasoning → Reasoning benchmarks and evaluation
MiroThinker: Pushing the Performance Boundaries of Open-Source Research Agents via Model, Context, and Interactive ScalingLLM reasoning → Reasoning for agents, planning and code +1
From Single to Societal: Analyzing Persona-Induced Bias in Multi-Agent InteractionsLLM agents and tool use → Social simulation and multi-agent debate
MALBO: Optimizing LLM-Based Multi-Agent Teams via Multi-Objective Bayesian OptimizationLLM agents and tool use → General LLM agent frameworks and benchmarks
NegBLEURT Forest: Leveraging Inconsistencies for Detecting Jailbreak AttacksLLM safety and security → Jailbreaking
Scaling Equitable Reflection Assessment in Education via Large Language Models and Role-Based Feedback AgentsLLM agents and tool use → Educational agents and math problem solving
Learning to Refine: An Agentic RL Approach for Iterative SPARQL Query ConstructionLLM agents and tool use → Web, search, and research agents
Socrates-Mol: Self-Oriented Cognitive Reasoning through Autonomous Trial-and-Error with Empirical-Bayesian Screening for MoleculesLLM reasoning → Clinical and biomedical applications
Drone Swarm Energy ManagementDeep reinforcement learning methods → RL for robot navigation and UAV control
DocLens : A Tool-Augmented Multi-Agent Framework for Long Visual Document UnderstandingMultimodal LLMs / vision-language models → Document understanding and multimodal RAG +1
Aligning Machiavellian Agents: Behavior Steering via Test-Time Policy ShapingLLM alignment / RLHF / preference optimisation → Human values and AI feedback alignment
Bridging Hidden States in Vision-Language ModelsMultimodal LLMs / vision-language models → VLM benchmarks and visual reasoning
Scalable Policy Evaluation with Video World ModelsWorld models → Embodied and robot world models +1
Experience-Guided Adaptation of Inference-Time Reasoning StrategiesLLM reasoning → Reasoning for agents, planning and code +1
W2S-AlignTree: Weak-to-Strong Inference-Time Alignment for Large Language Models via Monte Carlo Tree SearchLLM alignment / RLHF / preference optimisation → LLM preference optimization methods
FarSkip-Collective: Unhobbling Blocking Communication in Mixture of Experts ModelsMixture-of-experts → Efficient MoE training and inference systems
PAS : Prelim Attention Score for Detecting Object Hallucinations in Large Vision--Language ModelsMultimodal LLMs / vision-language models → Hallucination mitigation in LVLMs
Intrinsic Dimension Estimation for Radio Galaxy Zoo using Diffusion ModelsDiffusion models → Inverse problems and image restoration
Rethinking Progression of Memory State in Robotic Manipulation: An Object-Centric PerspectiveVision-language-action models → Reasoning and chain-of-thought in VLA
Rethinking Efficient Mixture-of-Experts for Remote Sensing Modality-Missing ClassificationMixture-of-experts → Vision mixture-of-experts
VoxTell: Free-Text Promptable Universal 3D Medical Image SegmentationMultimodal LLMs / vision-language models → Medical vision-language models
DiffPro: Joint Timestep and Layer-Wise Precision Optimization for Efficient Diffusion InferenceDiffusion models → Diffusion foundations and sampling
Synthetic Stimuli, Real Gains: Rethinking VLM Fine-Tuning Through Fully Controlled Data GenerationLLM reasoning → Multimodal reasoning +1
VP-Bench: A Comprehensive Benchmark for Visual Prompting in Multimodal Large Language ModelsMultimodal LLMs / vision-language models → MLLM training and visual instruction tuning
The Persistence of Cultural Memory: Investigating Multimodal Iconicity in Diffusion ModelsDiffusion models → Image generation, text-to-image and editing
Low-Bit, High-Fidelity: Optimal Transport Quantization for Flow MatchingDiffusion models → Flow matching, time series and science
Q-Doc: Benchmarking Document Image Quality Assessment Capabilities in Multi-modal Large Language ModelsMultimodal LLMs / vision-language models → Document understanding and multimodal RAG
MicroVQA++: High-Quality Microscopy Reasoning Dataset with Weakly Supervised Graphs for Multimodal Large Language ModelLLM reasoning → Clinical and biomedical applications +1
Multi-Phase Spacecraft Trajectory Optimization via Transformer-Based Reinforcement LearningDeep reinforcement learning methods → RL for robot navigation and UAV control
MarsRL: Advancing Multi-Agent Reasoning System via Reinforcement Learning with Agentic Pipeline ParallelismLLM reasoning → Reasoning for agents, planning and code +1
SRLF: An Agent-Driven Set-Wise Reflective Learning Framework for Sequential RecommendationLLM agents and tool use → Agentic recommender systems
On-Device Fine-Tuning via Backprop-Free Zeroth-Order OptimizationLLM agents and tool use → Agent memory systems
SEAL: Subspace-Anchored Watermarks for LLM OwnershipLLM safety and security → Watermarking generated content and models
Privacy Challenges and Solutions in Retrieval-Augmented Generation-Enhanced LLMs for Healthcare Chatbots: A Review of Applications, Risks, and Future DirectionsRetrieval-augmented generation → Privacy and security of RAG
UFO3: Weaving the Digital Agent GalaxyLLM agents and tool use → General LLM agent frameworks and benchmarks
DocSLM: A Small Vision-Language Model for Long Multimodal Document UnderstandingLLM reasoning → Multimodal reasoning +1
iMAD: Intelligent Multi-Agent Debate for Efficient and Accurate LLM InferenceLLM agents and tool use → Social simulation and multi-agent debate
EcoAlign: An Economically Rational Framework for Efficient LVLM AlignmentLLM reasoning → Reasoning for agents, planning and code +2
AUVIC: Adversarial Unlearning of Visual Concepts for Multi-modal Large Language ModelsMultimodal LLMs / vision-language models → MLLM jailbreaks and adversarial robustness +1
Experiences from Benchmarking Vision-Language-Action Models for Robotic ManipulationVision-language-action models → Generalist VLA models and benchmarks
GraphPilot: Grounded Scene Graph Conditioning for Language-Based Autonomous DrivingMultimodal LLMs / vision-language models → Driving VLMs and open-vocabulary detection +1
Discovering Meaningful Units with Visually Grounded Semantics from Image CaptionsMultimodal LLMs / vision-language models → VLM benchmarks and visual reasoning
AIonopedia: an LLM agent orchestrating multimodal learning for ionic liquid discoveryLLM agents and tool use → Medical and scientific domain agents
Align3GR: Unified Multi-Level Alignment for LLM-based Generative RecommendationLLM reasoning → LLM reasoning for recommendation
CountSteer: Steering Attention for Object Counting in Diffusion ModelsDiffusion models → Image generation, text-to-image and editing
UAVBench: An Open Benchmark Dataset for Autonomous and Agentic AI UAV Systems via LLM-Generated Flight ScenariosLLM reasoning → Spatial and embodied reasoning +1
Arcee: Differentiable Recurrent State Chain for Generative Vision Modeling with Mamba SSMsState-space sequence models (S4, Mamba) → Vision Mamba backbones
Beyond Flatlands: Unlocking Spatial Intelligence by Decoupling 3D Reasoning from Numerical RegressionLLM reasoning → Spatial and embodied reasoning +1
StyleQoRA: Quality-Aware Low-Rank Adaptation for Few-Shot Multi-Style EditingMixture-of-experts → Mixture of LoRA experts
STaR: Towards Effective and Stable Table Reasoning via Slow-Thinking Large Language ModelsLLM reasoning → Knowledge- and tool-augmented reasoning
DoReMi: Bridging 3D Domains via Topology-Aware Domain-Representation Mixture of ExpertsMixture-of-experts → Vision mixture-of-experts
Humanoid Whole-Body Badminton via an Annealed Reinforcement Learning CurriculumDeep reinforcement learning methods → RL for robot locomotion and manipulation
Questioning the Stability of Visual Question AnsweringMultimodal LLMs / vision-language models → VLM benchmarks and visual reasoning
Geospatial Chain of Thought Reasoning for Enhanced Visual Question Answering on Satellite ImageryLLM reasoning → Spatial and embodied reasoning
LoRaCompass: Robust Reinforcement Learning to Efficiently Search for a LoRa TagDeep reinforcement learning methods → RL for robot locomotion and manipulation
ReTrace: Interactive Visualizations for Reasoning Traces of Large Reasoning ModelsLLM reasoning → Reasoning benchmarks and evaluation
Multi-agent Undercover Gaming: Hallucination Removal via Counterfactual Test for Multimodal ReasoningLLM reasoning → Hallucination detection and mitigation +1
Viper-F1: Fast and Fine-Grained Multimodal Understanding with Cross-Modal State-Space ModulationMultimodal LLMs / vision-language models → MLLM training and visual instruction tuning
Refine and Align: Confidence Calibration through Multi-Agent Interaction in VQALLM agents and tool use → Multimodal and embodied agents
From Hindsight to Foresight: Self-Encouraged Hindsight Distillation for Knowledge-based Visual Question AnsweringLLM reasoning → Multimodal reasoning
Convergence of Flow-Policy Gradient Learning for Linear Quadratic Regulator ProblemsOffline RL and imitation learning from data → Diffusion and flow policies for offline RL +1
VIDEOP2R: Video Understanding from Perception to ReasoningLLM reasoning → Multimodal reasoning
AccKV: Towards Efficient Audio-Video LLMs Inference via Adaptive-Focusing and Cross-Calibration KV Cache OptimizationEfficient LLM inference and serving → KV cache optimization
Efficient Reinforcement Learning for Zero-Shot Coordination in Evolving GamesDeep reinforcement learning methods → RL for games and combinatorial optimization
ARCTraj: A Dataset and Benchmark of Human Reasoning Trajectories for Abstract Problem SolvingLLM reasoning → Multimodal reasoning
Evaluating Latent Generative Paradigms for High-Fidelity 3D Shape Completion from a Single Depth ImageDiffusion models → 3D, point cloud and scene generation
LiteAttention: A Temporal Sparse Attention for Diffusion TransformersDiffusion models → Video generation
Autonomous Vehicle Path Planning by Searching With Differentiable SimulationEnd-to-end autonomous driving → Imitation-learning end-to-end planners
Key Decision-Makers in Multi-Agent Debates: Who Holds the Power?LLM reasoning → Reasoning for agents, planning and code +1
GraphMASAL: A Graph-based Multi-Agent System for Adaptive LearningLLM agents and tool use → Educational agents and math problem solving
CrossMed: A Multimodal Cross-Task Benchmark for Compositional Generalization in Medical ImagingMultimodal LLMs / vision-language models → Medical vision-language models
AirCopBench: A Benchmark for Multi-drone Collaborative Embodied Perception and ReasoningLLM reasoning → Spatial and embodied reasoning +1
Data Poisoning Vulnerabilities Across Healthcare AI Architectures: A Security Threat AnalysisLLM safety and security → Backdoor and data poisoning attacks

All 403