Papers

New on arXiv

Thu 13 Nov 2025

409 papers

PlotGen-Bench: Evaluating VLMs on Generating Visualization Code from Diverse Plots across Multiple LibrariesMultimodal LLMs / vision-language models → Chart understanding and anomaly detection
LM4Opt-RA: A Multi-Candidate LLM Framework with Structured Ranking for Automating Network Resource AllocationLLM reasoning → Test-time compute and efficient thinking
A3: Attention-Aware Accurate KV Cache Fusion for Fast Large Language Model ServingRetrieval-augmented generation → Generation and context integration +1
SCALEX: Scalable Concept and Latent Exploration for Diffusion ModelsDiffusion models → Image generation, text-to-image and editing
Can AI Models be Jailbroken to Phish Elderly Victims? An End-to-End EvaluationLLM safety and security → Prompt injection and agent security
Towards autonomous quantum physics research using LLM agents with access to intelligent toolsLLM agents and tool use → Web, search, and research agents
Concept-RuleNet: Grounded Multi-Agent Neurosymbolic Reasoning in Vision Language ModelsLLM reasoning → Multimodal reasoning +2
Diffusion Models: A Mathematical IntroductionDiffusion models → Diffusion foundations and sampling
Uncertainty Makes It Stable: Curiosity-Driven Quantized Mixture-of-ExpertsMixture-of-experts → MoE large language models
ExpertAD: Enhancing Autonomous Driving Systems with Mixture of ExpertsEnd-to-end autonomous driving → Reasoning VLA and VLM driving policies +1
Speculative Decoding in Decentralized LLM Inference: Turning Communication Latency into Computation ThroughputEfficient LLM inference and serving → Speculative decoding
Exposing DeepFakes via Hyperspectral Domain MappingDiffusion models → Inverse problems and image restoration
Optimizing Input of Denoising Score Matching is Biased Towards Higher Score NormDiffusion models → Diffusion foundations and sampling
HPCAgentTester: A Multi-Agent LLM Approach for Enhanced HPC Unit Test GenerationLLM agents and tool use → Software engineering coding agents
Enhancing Demand-Oriented Regionalization with Agentic AI and Local Heterogeneous Data for Adaptation PlanningLLM agents and tool use → Multimodal and embodied agents
Leveraging Parameter Space Symmetries for Reasoning Skill Transfer in LLMsLLM reasoning → Test-time compute and efficient thinking
HARNESS: Human-Agent Risk Navigation and Event Safety System for Proactive Hazard Forecasting in High-Risk DOE EnvironmentsLLM agents and tool use → Agent threats, guardrails and governance
Curved Worlds, Clear Boundaries: Generalizing Speech Deepfake Detection using Hyperbolic and Spherical Geometry SpacesDiffusion models → Audio, speech and music
From Efficiency to Adaptivity: A Deeper Look at Adaptive Reasoning in Large Language ModelsLLM reasoning → Test-time compute and efficient thinking
Fast Data Attribution for Text-to-Image ModelsDiffusion models → Diffusion foundations and sampling
PISanitizer: Preventing Prompt Injection to Long-Context LLMs via Prompt SanitizationLLM safety and security → Prompt injection and agent security
BadThink: Triggered Overthinking Attacks on Chain-of-Thought Reasoning in Large Language ModelsLLM reasoning → Reasoning for agents, planning and code
Bias-Restrained Prefix Representation Finetuning for Mathematical ReasoningLLM reasoning → Mathematical reasoning
Co-EPG: A Framework for Co-Evolution of Planning and Grounding in Autonomous GUI AgentsLLM agents and tool use → GUI and mobile agents
Enhancing the Outcome Reward-based RL Training of MLLMs with Self-Consistency SamplingLLM reasoning → Multimodal reasoning +1
ParoQuant: Pairwise Rotation Quantization for Efficient Reasoning LLM InferenceLLM reasoning → Test-time compute and efficient thinking
One Small Step in Latent, One Giant Leap for Pixels: Fast Latent Upscale Adapter for Your Diffusion ModelsDiffusion models → Diffusion foundations and sampling
SSR: Socratic Self-Refine for Large Language Model ReasoningLLM reasoning → Test-time compute and efficient thinking
Towards Blind and Low-Vision Accessibility of Lightweight VLMs and Custom LLM-EvalsMultimodal LLMs / vision-language models → VLM benchmarks and visual reasoning
Towards an Agentic Workflow for Internet Measurement ResearchLLM agents and tool use → Software engineering coding agents
Multitask GLocal OBIA-Mamba for Sentinel-2 Landcover MappingState-space sequence models (S4, Mamba) → Point clouds, detection and remote sensing
URaG: Unified Retrieval and Generation in Multimodal LLMs for Efficient Long Document UnderstandingMultimodal LLMs / vision-language models → Document understanding and multimodal RAG
Convomem Benchmark: Why Your First 150 Conversations Don't Need RAGRetrieval-augmented generation → Conversational and speech RAG
Say It Differently: Linguistic Styles as Jailbreak VectorsLLM safety and security → Jailbreaking
SemanticVLA: Semantic-Aligned Sparsification and Enhancement for Efficient Robotic ManipulationVision-language-action models → Efficient VLA inference and adaptation
Mastering Olympiad-Level Physics with Artificial IntelligenceLLM reasoning → Spatial and embodied reasoning
LocalBench: Benchmarking LLMs on County-Level Local Knowledge and ReasoningLLM reasoning → Spatial and embodied reasoning
Reasoning about Intent for Ambiguous RequestsLLM reasoning → Reasoning benchmarks and evaluation
Enhanced Privacy Leakage from Noise-Perturbed Gradients via Gradient-Guided Conditional Diffusion ModelsDiffusion models → Robustness, privacy and adversarial aspects
Generalizing Analogical Inference from Boolean to Continuous DomainsLLM reasoning → Reasoning benchmarks and evaluation
nuPlan-R: A Closed-Loop Planning Benchmark for Autonomous Driving via Reactive Multi-Agent SimulationEnd-to-end autonomous driving → Imitation-learning end-to-end planners
Rethinking the Reliability of Multi-agent System: A Perspective from Byzantine Fault ToleranceLLM reasoning → Reasoning for agents, planning and code +1
AgentEvolver: Towards Efficient Self-Evolving Agent SystemLLM agents and tool use → General LLM agent frameworks and benchmarks +1
MonkeyOCR v1.5 Technical Report: Unlocking Robust Document Parsing for Complex PatternsRetrieval-augmented generation → Multimodal RAG
Operator Models for Continuous-Time Offline Reinforcement LearningOffline RL and imitation learning from data → Offline RL algorithms
Position: On the Methodological Pitfalls of Evaluating Base LLMs for ReasoningLLM reasoning → Reasoning benchmarks and evaluation
TruthfulRAG: Resolving Factual-level Conflicts in Retrieval-Augmented Generation with Knowledge GraphsRetrieval-augmented generation → RAG robustness and adaptive retrieval
SITA: A Framework for Structure-to-Instance Theorem AutoformalizationLLM reasoning → Mathematical reasoning
BhashaKritika: Building Synthetic Pretraining Data at Scale for Indic LanguagesSynthetic data and self-improvement for LLMs → LLM synthetic data generation
Rectify Evaluation Preference: Improving LLMs' Critique on Math Reasoning via Perplexity-aware Reinforcement LearningLLM reasoning → Mathematical reasoning
Rethinking Visual Information Processing in Multimodal LLMsLLM reasoning → Multimodal reasoning +1
Generalizable Slum Detection from Satellite Imagery with Mixture-of-ExpertsMixture-of-experts → Vision mixture-of-experts
Local Hybrid Retrieval-Augmented Document QARetrieval-augmented generation → RAG robustness and adaptive retrieval
Adaptive Residual-Update Steering for Low-Overhead Hallucination Mitigation in Large Vision Language ModelsMultimodal LLMs / vision-language models → Hallucination mitigation in LVLMs
OutSafe-Bench: A Benchmark for Multimodal Offensive Content Detection in Large Language ModelsMultimodal LLMs / vision-language models → MLLM jailbreaks and adversarial robustness
Modeling User-System Behavior for Training-free Building of Private Domain Conversational AgentsLLM agents and tool use → General LLM agent frameworks and benchmarks
PROPA: Toward Process-level Optimization in Visual Reasoning via Reinforcement LearningLLM reasoning → Multimodal reasoning +1
Causal-HalBench: Uncovering LVLMs Object Hallucinations Through Causal InterventionMultimodal LLMs / vision-language models → Hallucination mitigation in LVLMs
Facial-R1: Aligning Reasoning and Recognition for Facial Emotion AnalysisLLM reasoning → Audio-language model reasoning
ProgRAG: Hallucination-Resistant Progressive Retrieval and Reasoning over Knowledge GraphsLLM reasoning → Knowledge- and tool-augmented reasoning
Lost in Serialization: Invariance and Generalization of LLM Graph ReasonersLLM reasoning → Knowledge- and tool-augmented reasoning
VocalNet-M2: Advancing Low-Latency Spoken Language Modeling via Integrated Multi-Codebook Tokenization and Multi-Token PredictionDiffusion models → Audio, speech and music
Speech-Audio Compositional Attacks on Multimodal LLMs and Their Mitigation with SALMONN-GuardLLM safety and security → Audio jailbreaks and speech backdoors
EffiReason-Bench: A Unified Benchmark for Evaluating and Advancing Efficient Reasoning in Large Language ModelsLLM reasoning → Test-time compute and efficient thinking
Improved Offline Reinforcement Learning via Quantum Metric EncodingOffline RL and imitation learning from data → Offline RL algorithms
Beyond the Black Box: Demystifying Multi-Turn LLM Reasoning with VISTALLM reasoning → Reasoning benchmarks and evaluation
RAGFort: Dual-Path Defense Against Proprietary Knowledge Base Extraction in Retrieval-Augmented GenerationRetrieval-augmented generation → Privacy and security of RAG
Learning a Thousand Tasks in a DayOffline RL and imitation learning from data → Robot manipulation imitation learning
MTAttack: Multi-Target Backdoor Attacks against Large Vision-Language ModelsMultimodal LLMs / vision-language models → MLLM jailbreaks and adversarial robustness +1
Opinion: Towards Unified Expressive Policy Optimization for Robust Robot LearningOffline RL and imitation learning from data → Real-world robot offline RL
GridPrune: From "Where to Look" to "What to Select" in Visual Token Pruning for MLLMsMultimodal LLMs / vision-language models → Visual token pruning and efficiency +1
Format Matters: The Robustness of Multimodal LLMs in Reviewing Evidence from Tables and ChartsMultimodal LLMs / vision-language models → Chart understanding and anomaly detection
RadHiera: Semantic Hierarchical Reinforcement Learning for Medical Report GenerationMultimodal LLMs / vision-language models → Medical vision-language models
When Eyes and Ears Disagree: Can MLLMs Discern Audio-Visual Confusion?Multimodal LLMs / vision-language models → Video understanding with MLLMs
Physical Plausibility Reasoning via HCM-GRPO: Empowering Compact Model for Superior PerformanceLLM reasoning → Multimodal reasoning
BuddyMoE: Exploiting Expert Redundancy to Accelerate Memory-Constrained Mixture-of-Experts InferenceMixture-of-experts → Efficient MoE training and inference systems
Continuous Benchmark Generation for Evaluating Enterprise-scale LLM AgentsLLM agents and tool use → General LLM agent frameworks and benchmarks
Efficient Thought Space Exploration Through Strategic InterventionLLM reasoning → Test-time compute and efficient thinking
Beyond ReAct: A Planner-Centric Framework for Complex Tool-Augmented LLM ReasoningLLM reasoning → Reasoning for agents, planning and code +1
Moral Change or Noise? On Problems of Aligning AI With Temporally Unstable Human FeedbackLLM alignment / RLHF / preference optimisation → Human values and AI feedback alignment
AffordBot: 3D Fine-grained Embodied Reasoning via Multimodal Large Language ModelsLLM reasoning → Spatial and embodied reasoning +1
fastbmRAG: A Fast Graph-Based RAG Framework for Efficient Processing of Large-Scale Biomedical LiteratureRetrieval-augmented generation → Medical and biomedical RAG
Phantom Menace: Exploring and Enhancing the Robustness of VLA Models Against Physical Sensor AttacksVision-language-action models → Safety and adversarial attacks
SPAN: Benchmarking and Improving Cross-Calendar Temporal Reasoning of Large Language ModelsLLM reasoning → Knowledge- and tool-augmented reasoning
Language Drift in Multilingual Retrieval-Augmented Generation: Characterization and Decoding-Time MitigationRetrieval-augmented generation → Generation and context integration
Modeling Uncertainty Trends for Timely Retrieval in Dynamic RAGRetrieval-augmented generation → RAG robustness and adaptive retrieval
Difference Vector Equalization for Robust Fine-tuning of Vision-Language ModelsMultimodal LLMs / vision-language models → Prompt learning and adaptation of CLIP
REAP: Enhancing RAG with Recursive Evaluation and Adaptive Planning for Multi-Hop Question AnsweringRetrieval-augmented generation → Knowledge-graph and structured-knowledge RAG
Equivariant Sampling for Improving Diffusion Model-based Image RestorationDiffusion models → Inverse problems and image restoration
Audio-VLA: Adding Contact Audio Perception to Vision-Language-Action Model for Robotic ManipulationVision-language-action models → Generalist VLA models and benchmarks
Robust Object Detection with Pseudo Labels from VLMs using Per-Object Co-teachingMultimodal LLMs / vision-language models → Driving VLMs and open-vocabulary detection
MosaicDoc: A Large-Scale Bilingual Benchmark for Visually Rich Document UnderstandingMultimodal LLMs / vision-language models → Document understanding and multimodal RAG
MINDS: A Cross-cultural Dialogue Corpus for Social Norm Classification and Adherence DetectionRetrieval-augmented generation → Conversational and speech RAG
Learning to Pose Problems: Reasoning-Driven and Solver-Adaptive Data SynthesisLLM reasoning → Mathematical reasoning +1
Simulator and Experience Enhanced Diffusion Model for Comprehensive ECG GenerationDiffusion models → Medical imaging and segmentation
HCC-3D: Hierarchical Compensatory Compression for 98% 3D Token Reduction in Vision-Language ModelsLLM reasoning → Spatial and embodied reasoning +2
EnchTable: Unified Safety Alignment Transfer in Fine-tuned Large Language ModelsLLM alignment / RLHF / preference optimisation → Safety alignment and jailbreak attacks +1
Remember Me: Bridging the Long-Range Gap in LVLMs with Three-Step Inference-Only Decay Resilience StrategiesMultimodal LLMs / vision-language models → Visual token pruning and efficiency
In-Token Rationality Optimization: Towards Accurate and Concise LLM Reasoning via Self-FeedbackLLM reasoning → Test-time compute and efficient thinking
Uncertainty-Guided Checkpoint Selection for Reinforcement Finetuning of Large Language ModelsLLM alignment / RLHF / preference optimisation → Reward modeling and RLHF

All 409