Papers

New on arXiv

Wed 26 Mar 2025

329 papers

3MDBench: Medical Multimodal Multi-agent Dialogue BenchmarkMultimodal LLMs / vision-language models → Medical vision-language models +1
A Survey of Multimodal Retrieval-Augmented GenerationRetrieval-augmented generation → Multimodal RAG
Adaptive State-Space Mamba for Real-Time Sensor Data Anomaly DetectionState-space sequence models (S4, Mamba) → Time series forecasting with SSMs
ShieldAgent: Shielding Agents via Verifiable Safety Policy ReasoningLLM reasoning → Reasoning for agents, planning and code +2
Reasoning Beyond Limits: Advances and Open Problems for LLMsLLM reasoning → Reasoning benchmarks and evaluation +1
Ancestral Mamba: Enhancing Selective Discriminant Space Model with Online Visual Prototype Learning for Efficient and Robust Discriminant ApproachState-space sequence models (S4, Mamba) → Vision Mamba backbones
InfoBid: A Simulation Framework for Studying Information Disclosure in Auctions with Large Language Model-based AgentsLLM agents and tool use → Financial trading and forecasting agents
A Spatial-temporal Deep Probabilistic Diffusion Model for Reliable Hail Nowcasting with Radar Echo ExtrapolationDiffusion models → Flow matching, time series and science
Zero-Shot LLMs in Human-in-the-Loop RL: Replacing Human Feedback for Reward ShapingLLM alignment / RLHF / preference optimisation → Reward modeling and RLHF
Protecting Your Video Content: Disrupting Automated Video-based LLM AnnotationsLLM safety and security → Watermarking generated content and models
Low-Rank Adaptation of Pre-Trained Stable Diffusion for Rigid-Body Target ISAR ImagingDiffusion models → Inverse problems and image restoration
Optimizing Safe and Aligned Language Generation: A Multi-Objective GRPO ApproachLLM reasoning → Reinforcement learning for reasoning +1
Skip-Vision: Efficient and Scalable Acceleration of Vision-Language Models via Adaptive Token SkippingMultimodal LLMs / vision-language models → Visual token pruning and efficiency +1
Operating Room Workflow Analysis via Reasoning Segmentation over Digital TwinsLLM reasoning → Clinical and biomedical applications
World Model Agents with Change-Based Intrinsic MotivationWorld models → Latent dynamics models and model-based RL +1
The Art of Tool Interface DesignLLM agents and tool use → Tool learning and function calling
ENCORE: Entropy-guided Reward Composition for Multi-head Safety Reward ModelsLLM alignment / RLHF / preference optimisation → Safety alignment and jailbreak attacks
Reinforcement Learning for Efficient Toxicity Detection in Competitive Online Video GamesDeep reinforcement learning methods → RL for games and combinatorial optimization
DEMENTIA-PLAN: An Agent-Based Framework for Multi-Knowledge Graph Retrieval-Augmented Generation in Dementia CareRetrieval-augmented generation → Medical and biomedical RAG +1
TransDiffSBDD: Causality-Aware Multi-Modal Structure-Based Drug DesignDiffusion models → Flow matching, time series and science
Unified Multimodal Discrete DiffusionDiffusion models → Image generation, text-to-image and editing
Robust Deep Reinforcement Learning in Robotics via Adaptive Gradient-Masked Adversarial AttacksDeep reinforcement learning methods → Adversarial robustness and RL in finance
TAR: Teacher-Aligned Representations via Contrastive Learning for Quadrupedal LocomotionKnowledge distillation → On-policy and self-distillation for LLMs +1
Debiasing Kernel-Based Generative ModelsDiffusion models → Diffusion foundations and sampling
Playing the Fool: Jailbreaking LLMs and Multimodal LLMs with Out-of-Distribution StrategyLLM alignment / RLHF / preference optimisation → Safety alignment and jailbreak attacks +1
FB-4D: Spatial-Temporal Coherent Dynamic 3D Content Generation with Feature BanksDiffusion models → 3D, point cloud and scene generation
Understanding R1-Zero-Like Training: A Critical PerspectiveLLM reasoning → Reinforcement learning for reasoning
MCTS-RAG: Enhancing Retrieval-Augmented Generation with Monte Carlo Tree SearchLLM reasoning → Knowledge- and tool-augmented reasoning +1
ADS-Edit: A Multimodal Knowledge Editing Dataset for Autonomous Driving SystemsMultimodal LLMs / vision-language models → Driving VLMs and open-vocabulary detection
Reason-RFT: Reinforcement Fine-Tuning for Visual Reasoning of Vision Language ModelsLLM reasoning → Multimodal reasoning +1
Optimal Scaling Laws for Efficiency Gains in a Theoretical Transformer-Augmented Sectional MoE FrameworkMixture-of-experts → Expert routing strategies
Can LLM Agents Simulate Multi-Turn Human Behavior? Evidence from Real Online Customer Behavior DataLLM agents and tool use → Social simulation and multi-agent debate
Math Blind: Failures in Diagram Understanding Undermine Reasoning in MLLMsLLM reasoning → Multimodal reasoning
High Quality Diffusion Distillation on a Single GPU with Relative and Absolute Position MatchingDiffusion models → Diffusion foundations and sampling
RecTable: Fast Modeling Tabular Data with Rectified FlowDiffusion models → Flow matching, time series and science
Learning Straight Flows by Learning Curved InterpolantsDiffusion models → Flow matching, time series and science
From Annotation to Adaptation: Metrics, Synthetic Data, and Aspect Extraction for Aspect-Based Sentiment Analysis with Large Language ModelsSynthetic data and self-improvement for LLMs → LLM synthetic data generation
MMMORRF: Multimodal Multilingual Modularized Reciprocal Rank FusionMultimodal LLMs / vision-language models → Video understanding with MLLMs
Graph-Enhanced Model-Free Reinforcement Learning Agents for Efficient Power Grid Topological ControlDeep reinforcement learning methods → RL for energy, networks and operations
GLRD: Global-Local Collaborative Reason and Debate with PSL for 3D Open-Vocabulary DetectionLLM reasoning → Spatial and embodied reasoning
Mitigating Low-Level Visual Hallucinations Requires Self-Awareness: Database, Model and Training StrategyMultimodal LLMs / vision-language models → Hallucination mitigation in LVLMs
TAMA: A Human-AI Collaborative Thematic Analysis Framework Using Multi-Agent LLMs for Clinical InterviewsLLM agents and tool use → Medical and scientific domain agents
AutoRad-Lung: A Radiomic-Guided Prompting Autoregressive Vision-Language Model for Lung Nodule Malignancy PredictionMultimodal LLMs / vision-language models → Medical vision-language models
AccidentSim: Generating Vehicle Collision Videos with Physically Realistic Collision Trajectories from Real-World Accident ReportsWorld models → World models for autonomous driving
MMGen: Unified Multi-modal Image Generation and Understanding in One GoDiffusion models → Image generation, text-to-image and editing
Unlocking Efficient Long-to-Short LLM Reasoning with Model MergingLLM reasoning → Test-time compute and efficient thinking
Enhancing Multi-modal Models with Heterogeneous MoE Adapters for Fine-tuningMixture-of-experts → Multimodal MoE models
State-Aware Perturbation Optimization for Robust Deep Reinforcement LearningDeep reinforcement learning methods → Adversarial robustness and RL in finance
IAP: Improving Continual Learning of Vision-Language Models via Instance-Aware PromptingMultimodal LLMs / vision-language models → Prompt learning and adaptation of CLIP
Diffusion Counterfactuals for Image RegressorsDiffusion models → Diffusion foundations and sampling
What to Retrieve for Effective Retrieval-Augmented Code Generation? An Empirical Study and BeyondRetrieval-augmented generation → RAG for code
Synthetic Data Augmentation for Cross-domain Implicit Discourse Relation RecognitionSynthetic data and self-improvement for LLMs → LLM synthetic data generation
Optimizing Case-Based Reasoning System for Functional Test Script Generation with Large Language ModelsLLM reasoning → Reasoning for agents, planning and code
Exploring Robustness of Cortical Morphometry in the presence of white matter lesions, using Diffusion Models for Lesion FillingDiffusion models → Medical imaging and segmentation
Injecting Adrenaline into LLM Serving: Boosting Resource Utilization and Throughput via Attention DisaggregationEfficient LLM inference and serving → Serving systems and request scheduling
TD-BFR: Truncated Diffusion Model for Efficient Blind Face RestorationDiffusion models → Inverse problems and image restoration
Knowledge-Based Multi-Agent Framework for Automated Software Architecture DesignLLM agents and tool use → Software engineering coding agents
Accelerate Parallelizable Reasoning via Parallel Decoding within One SequenceLLM reasoning → Test-time compute and efficient thinking +1
StableToolBench-MirrorAPI: Modeling Tool Environments as Mirrors of 7,000+ Real-World APIsLLM agents and tool use → Tool learning and function calling
GAIA-2: A Controllable Multi-View Generative World Model for Autonomous DrivingWorld models → World models for autonomous driving +1
UniVRSE: Unified Vision-conditioned Response Semantic Entropy for Hallucination Detection in Medical Vision-Language ModelsMultimodal LLMs / vision-language models → Hallucination mitigation in LVLMs
MLLM-Selector: Necessity and Diversity-driven High-Value Data Selection for Enhanced Visual Instruction TuningMultimodal LLMs / vision-language models → MLLM training and visual instruction tuning
Towards Efficient and General-Purpose Few-Shot Misclassification Detection for Vision-Language ModelsMultimodal LLMs / vision-language models → Prompt learning and adaptation of CLIP
VPO: Aligning Text-to-Video Generation Models with Prompt OptimizationLLM alignment / RLHF / preference optimisation → Diffusion and multimodal model alignment
Contrastive Learning Guided Latent Diffusion Model for Image-to-Image TranslationDiffusion models → Image generation, text-to-image and editing
Dissecting and Mitigating Diffusion Bias via Mechanistic InterpretabilityDiffusion models → Image generation, text-to-image and editing +1
From Trial to Triumph: Advancing Long Video Understanding via Visual Context Sample Scaling and Self-reward AlignmentMultimodal LLMs / vision-language models → Video understanding with MLLMs +1
Multi-agent Uncertainty-Aware Pessimistic Model-Based Reinforcement Learning for Connected Autonomous VehiclesEnd-to-end autonomous driving → Cooperative and V2X driving
The Crucial Role of Problem Formulation in Real-World Reinforcement LearningDeep reinforcement learning methods → Safe and constrained RL
RALLRec+: Retrieval Augmented Large Language Model Recommendation with ReasoningLLM reasoning → LLM reasoning for recommendation +1
Latent Beam Diffusion Models for Generating Visual SequencesDiffusion models → Image generation, text-to-image and editing
Perspective-Shifted Neuro-Symbolic World Models: A Framework for Socially-Aware Robot NavigationWorld models → World models for navigation
ITA-MDT: Image-Timestep-Adaptive Masked Diffusion Transformer Framework for Image-Based Virtual Try-OnDiffusion models → 3D, point cloud and scene generation
MoLe-VLA: Dynamic Layer-skipping Vision Language Action Model via Mixture-of-Layers for Efficient Robot ManipulationMultimodal LLMs / vision-language models → Embodied and 3D vision-language models +2
Self-ReS: Self-Reflection in Large Vision-Language Models for Long Video UnderstandingMultimodal LLMs / vision-language models → Video understanding with MLLMs
VideoGEM: Training-free Action Grounding in VideosMultimodal LLMs / vision-language models → Video understanding with MLLMs
Dynamic Pyramid Network for Efficient Multimodal Large Language ModelMultimodal LLMs / vision-language models → Visual token pruning and efficiency
Iterative Prompting with Persuasion Skills in Jailbreaking Large Language ModelsLLM safety and security → Jailbreaking
Wan: Open and Advanced Large-Scale Video Generative ModelsWorld models → Video-generation world models and simulators
Instruction-Oriented Preference Alignment for Enhancing Multi-Modal Comprehension Capability of MLLMsMultimodal LLMs / vision-language models → MLLM training and visual instruction tuning +1
A Multilingual, Culture-First Approach to Addressing Misgendering in LLM ApplicationsLLM safety and security → Safety alignment, refusal and guardrails
Traversing Distortion-Perception Tradeoff using a Single Score-Based Generative ModelDiffusion models → Inverse problems and image restoration
QualiSpeech: A Speech Quality Assessment Dataset with Natural Language Reasoning and DescriptionsLLM reasoning → Audio-language model reasoning
Model-Based Offline Reinforcement Learning with Adversarial Data AugmentationOffline RL and imitation learning from data → Offline RL algorithms
Faster Parameter-Efficient Tuning with Token Redundancy ReductionEfficient LLM inference and serving → Early exit and edge inference
sudo rm -rf agentic_securityLLM safety and security → Jailbreaking
ViLBench: A Suite for Vision-Language Process Reward ModelingLLM reasoning → Multimodal reasoning +1
EGVD: Event-Guided Video Diffusion Model for Physically Realistic Large-Motion Frame InterpolationDiffusion models → Video generation
Mamba-3D as Masked Autoencoders for Accurate and Data-Efficient Analysis of Medical Ultrasound VideosState-space sequence models (S4, Mamba) → Mamba for medical image analysis
How Secure is Forgetting? Linking Machine Unlearning to Machine Learning AttacksLLM safety and security → Machine unlearning
Unconditional Priors Matter! Improving Conditional Generation of Fine-Tuned Diffusion ModelsDiffusion models → Diffusion foundations and sampling
Video Motion GraphsDiffusion models → Video generation
Qwen2.5-Omni Technical ReportLLM reasoning → Audio-language model reasoning
Generalized Phase Pressure Control Enhanced Reinforcement Learning for Traffic Signal ControlDeep reinforcement learning methods → RL for autonomous driving and traffic
Open Deep Search: Democratizing Search with Open-source Reasoning AgentsLLM reasoning → Knowledge- and tool-augmented reasoning +1
Beyond Words: Advancing Long-Text Image Generation via Multimodal Autoregressive ModelsDiffusion models → Image generation, text-to-image and editing
GAPO: Learning Preferential Prompt through Generative Adversarial Policy OptimizationLLM alignment / RLHF / preference optimisation → LLM preference optimization methods
Rethinking Vision-Language Model in Face Forensics: Multi-Modal Interpretable Forged Face DetectorMultimodal LLMs / vision-language models → MLLM jailbreaks and adversarial robustness
Offline Reinforcement Learning with Discrete Diffusion SkillsDiffusion models → Diffusion and flow policies for robotics +1
Look Before Leap: Look-Ahead Planning with Uncertainty in Reinforcement LearningDeep reinforcement learning methods → RL theory and sample complexity

All 329