Papers

New on arXiv

Mon 2 Jun 2025

499 papers

Inverse Design in Distributed Circuits Using Single-Step Reinforcement LearningDeep reinforcement learning methods → RL for energy, networks and operations
Out-of-Vocabulary Sampling Boosts Speculative DecodingEfficient LLM inference and serving → Speculative decoding
Medical World Model: Generative Simulation of Tumor Evolution for Treatment PlanningWorld models → Medical and surgical world models
Absorb and Converge: Provable Convergence Guarantee for Absorbing Discrete Diffusion ModelsDiffusion models → Diffusion foundations and sampling
LAM SIMULATOR: Advancing Data Generation for Large Action Model Training via Online Exploration and Trajectory FeedbackLLM agents and tool use → General LLM agent frameworks and benchmarks
Improving Knowledge Distillation Under Unknown Covariate Shift Through Confidence-Guided Data AugmentationKnowledge distillation → General knowledge distillation methods
ImpRAG: Retrieval-Augmented Generation with Implicit QueriesRetrieval-augmented generation → RAG pipeline design and evaluation
CoDial: Interpretable Task-Oriented Dialogue Systems Through Dialogue Flow AlignmentLLM alignment / RLHF / preference optimisation → LLM preference optimization methods +1
What Makes LLMs Effective Sequential Recommenders? A Study on Preference Intensity and Temporal ContextLLM alignment / RLHF / preference optimisation → Preference alignment for recommenders and RAG
Are Mamba-based Audio Foundation Models the Best Fit for Non-Verbal Emotion Recognition?State-space sequence models (S4, Mamba) → Speech and audio SSMs
SafeOR-Gym: A Benchmark Suite for Safe Reinforcement Learning Algorithms on Practical Operations Research ProblemsDeep reinforcement learning methods → Safe and constrained RL
Physics-Guided Motion Loss for Video Generation ModelDiffusion models → Video generation
From Street Views to Urban Science: Discovering Road Safety Factors with Multimodal Large Language ModelsMultimodal LLMs / vision-language models → Driving VLMs and open-vocabulary detection
VLCD: Vision-Language Contrastive Distillation for Accurate and Efficient Automatic Placenta AnalysisKnowledge distillation → Distillation for medical imaging
Diff2Flow: Training Flow Matching Models via Diffusion Model AlignmentDiffusion models → Flow matching, time series and science +1
Improving LLM-Generated Code Quality with GRPOLLM reasoning → Reasoning for agents, planning and code
KDRL: Post-Training Reasoning LLMs via Unified Knowledge Distillation and Reinforcement LearningLLM reasoning → Reinforcement learning for reasoning +1
Reinforcement Learning with Data Bootstrapping for Dynamic Subgoal Pursuit in Humanoid Robot NavigationDeep reinforcement learning methods → RL for robot locomotion and manipulation
Bregman Centroid Guided Cross-Entropy MethodDeep reinforcement learning methods → RL theory and sample complexity
Act Only When It Pays: Efficient Reinforcement Learning for LLM Reasoning via Selective RolloutsLLM reasoning → Reinforcement learning for reasoning
HENT-SRT: Hierarchical Efficient Neural Transducer with Self-Distillation for Joint Speech Recognition and TranslationKnowledge distillation → Speech and audio distillation
Mitigating Data Poisoning Attacks to Local Differential PrivacyLLM safety and security → Membership inference attacks
Small Language Models are the Future of Agentic AILLM agents and tool use → General LLM agent frameworks and benchmarks
The Unified Cognitive Consciousness Theory for Language Models: Anchoring Semantics, Thresholds of Activation, and Emergent ReasoningLLM reasoning → Reasoning benchmarks and evaluation
Knowledge or Reasoning? A Close Look at How LLMs Think Across DomainsLLM reasoning → Clinical and biomedical applications
Hybrid AI for Responsive Multi-Turn Online Conversations with Novel Dynamic Routing and Feedback AdaptationRetrieval-augmented generation → Conversational and speech RAG
SynthRL: Scaling Visual Reasoning with Verifiable Data SynthesisLLM reasoning → Multimodal reasoning +2
Cycle Consistency as Reward: Learning Image-Text Alignment without Human PreferencesLLM alignment / RLHF / preference optimisation → Diffusion and multimodal model alignment
SALAD: Systematic Assessment of Machine Unlearning on LLM-Aided Hardware DesignLLM safety and security → Backdoor and data poisoning attacks
RATFM: Retrieval-augmented Time Series Foundation Model for Anomaly DetectionRetrieval-augmented generation → Time series and agricultural RAG
An Introduction to Flow Matching and Diffusion ModelsDiffusion models → Flow matching, time series and science
DRAG: Distilling RAG for SLMs from LLMs to Transfer Knowledge and Mitigate Hallucination via Evidence and Graph-based DistillationRetrieval-augmented generation → Hallucination detection in RAG
Fast-in-Slow: A Dual-System Foundation Model Unifying Fast Manipulation within Slow ReasoningLLM reasoning → Spatial and embodied reasoning
WebChoreArena: Evaluating Web Browsing Agents on Realistic Tedious Web TasksLLM agents and tool use → Web, search, and research agents
IMAGHarmony: Controllable Image Editing with Consistent Object Quantity and LayoutDiffusion models → Image generation, text-to-image and editing
3DRS: MLLMs Need 3D-Aware Representation Supervision for Scene UnderstandingLLM reasoning → Spatial and embodied reasoning +1
Learning Video Generation for Robotic Manipulation with Collaborative Trajectory ControlDiffusion models → Diffusion and flow policies for robotics
Beyond the 80/20 Rule: High-Entropy Minority Tokens Drive Effective Reinforcement Learning for LLM ReasoningLLM reasoning → Reinforcement learning for reasoning
RewardBench 2: Advancing Reward Model EvaluationLLM alignment / RLHF / preference optimisation → Reward modeling and RLHF
Image Generation from Contextually-Contradictory PromptsDiffusion models → Image generation, text-to-image and editing
Large language models can learn and generalize steganographic chain-of-thought under process supervisionLLM reasoning → Prompting and chain-of-thought reasoning
TaxaDiffusion: Progressively Trained Diffusion Model for Fine-Grained Species GenerationDiffusion models → Diffusion foundations and sampling
Unconditional CNN denoisers contain sparse semantic representation of imagesDiffusion models → Diffusion foundations and sampling
Reinforcement Learning Tuning for VideoLLMs: Reward Design and Data EfficiencyLLM reasoning → Multimodal reasoning +1
Enhancing Biomedical Multi-modal Representation Learning with Multi-scale Pre-training and Perturbed Report DiscriminationMultimodal LLMs / vision-language models → Medical vision-language models
COALESCE: Economic and Security Dynamics of Skill-Based Task Outsourcing Among Team of Autonomous LLM AgentsLLM agents and tool use → Agent skill learning and evolution
Agnostic Reinforcement Learning: Foundations and AlgorithmsDeep reinforcement learning methods → RL theory and sample complexity
CONFETTI: Conversational Function-Calling Evaluation Through Turn-Level InteractionsLLM agents and tool use → Tool learning and function calling
ShapeLLM-Omni: A Native Multimodal LLM for 3D Generation and UnderstandingMultimodal LLMs / vision-language models → Embodied and 3D vision-language models
MoDA: Modulation Adapter for Fine-Grained Visual Grounding in Instructional MLLMsMultimodal LLMs / vision-language models → MLLM training and visual instruction tuning
SmolVLA: A Vision-Language-Action Model for Affordable and Efficient RoboticsMultimodal LLMs / vision-language models → Embodied and 3D vision-language models +1
Beyond Static Responses: Multi-Agent LLM Systems as a New Paradigm for Social Science ResearchLLM agents and tool use → Social simulation and multi-agent debate
Backdoors in Code Summarizers: How Bad Is It?LLM safety and security → Backdoor and data poisoning attacks
Analysis of LLM Bias (Chinese Propaganda & Anti-US Sentiment) in DeepSeek-R1 vs. ChatGPT o3-mini-highLLM reasoning → Reasoning benchmarks and evaluation
A Study on the MCP x A2A Framework for Enhancing Interoperability of LLM-based Autonomous AgentsLLM agents and tool use → General LLM agent frameworks and benchmarks
OmniV2V: Versatile Video Generation and Editing via Dynamic Content ManipulationDiffusion models → Video generation
R2SM: Referring and Reasoning for Selective MasksLLM reasoning → Multimodal reasoning
iQUEST: An Iterative Question-Guided Framework for Knowledge Base Question AnsweringLLM reasoning → Knowledge- and tool-augmented reasoning
Harnessing Chain-of-Thought Reasoning in Multimodal Large Language Models for Face Anti-SpoofingLLM reasoning → Multimodal reasoning +1
DRAUN: An Algorithm-Agnostic Data Reconstruction Attack on Federated Unlearning SystemsLLM safety and security → Federated unlearning and poisoning
ReGA: Model-Based Safeguard for LLMs via Representation-Guided AbstractionLLM safety and security → Safety alignment, refusal and guardrails
ADEPT: Adaptive Diffusion Environment for Policy Transfer Sim-to-RealDiffusion models → Diffusion and flow policies for robotics +1
Many-for-Many: Unify the Training of Multiple Video and Image Generation and Manipulation TasksDiffusion models → Video generation
Data-assimilated model-informed reinforcement learningDeep reinforcement learning methods → RL theory and sample complexity
Thinking in Character: Advancing Role-Playing Agents with Role-Aware ReasoningLLM reasoning → Reasoning benchmarks and evaluation +1
VideoCap-R1: Enhancing MLLMs for Video Captioning via Structured ThinkingLLM reasoning → Multimodal reasoning +1
Active Learning via Vision-Language Model Adaptation with Open DataMultimodal LLMs / vision-language models → Prompt learning and adaptation of CLIP
Self-Challenging Language Model AgentsLLM agents and tool use → General LLM agent frameworks and benchmarks
SRPO: Enhancing Multimodal LLM Reasoning via Reflection-Aware Reinforcement LearningLLM reasoning → Multimodal reasoning +1
Reasoning-Table: Exploring Reinforcement Learning for Table ReasoningLLM reasoning → Knowledge- and tool-augmented reasoning
A Descriptive and Normative Theory of Human Beliefs in RLHFLLM alignment / RLHF / preference optimisation → Reward modeling and RLHF
Reasoning-Based Approach with Chain-of-Thought for Alzheimer's Detection Using Speech and Large Language ModelsLLM reasoning → Clinical and biomedical applications
K12Vista: Exploring the Boundaries of MLLMs in K-12 EducationLLM reasoning → Multimodal reasoning +1
MotionSight: Boosting Fine-Grained Motion Understanding in Multimodal LLMsMultimodal LLMs / vision-language models → Video understanding with MLLMs
Minimal Impact ControlNet: Advancing Multi-ControlNet IntegrationDiffusion models → Image generation, text-to-image and editing
EarthMind: Leveraging Cross-Sensor Data for Advanced Earth Observation Interpretation with a Unified Multimodal LLMMultimodal LLMs / vision-language models → Embodied and 3D vision-language models
Synthesis of discrete-continuous quantum circuits with multimodal diffusion modelsDiffusion models → Diffusion foundations and sampling
Leveraging Analytic Gradients in Provably Safe Reinforcement LearningDeep reinforcement learning methods → Safe and constrained RL
Zoom-Refine: Boosting High-Resolution Multimodal Understanding via Localized Zoom and Self-RefinementMultimodal LLMs / vision-language models → MLLM training and visual instruction tuning
Mixture of Experts Provably Detect and Learn the Latent Cluster Structure in Gradient-Based LearningMixture-of-experts → MoE theory and gating functions
Interpretable reinforcement learning for heat pump control through asymmetric differentiable decision treesDeep reinforcement learning methods → RL for energy, networks and operations
Bidirectional Soft Actor-Critic: Leveraging Forward and Reverse KL Divergence for Efficient Reinforcement LearningDeep reinforcement learning methods → RL theory and sample complexity
A Hierarchical Bin Packing Framework with Dual Manipulators via Heuristic Search and Deep Reinforcement LearningDeep reinforcement learning methods → RL for robot navigation and UAV control
Social Cooperation in Conversational AI AgentsLLM agents and tool use → General LLM agent frameworks and benchmarks
MAGIK: Mapping to Analogous Goals via Imagination-enabled Knowledge TransferLLM reasoning → Reinforcement learning for reasoning
General agents contain world modelsWorld models → World models for language agents
MLA-Trust: Benchmarking Trustworthiness of Multimodal LLM Agents in GUI EnvironmentsLLM agents and tool use → Agent threats, guardrails and governance
IndicRAGSuite: Large-Scale Datasets and a Benchmark for Indian Language RAG SystemsRetrieval-augmented generation → RAG pipeline design and evaluation
WoMAP: World Models For Embodied Open-Vocabulary Object LocalizationWorld models → Embodied and robot world models
Policy Newton Algorithm in Reproducing Kernel Hilbert SpaceDeep reinforcement learning methods → RL theory and sample complexity
Silence is Golden: Leveraging Adversarial Examples to Nullify Audio Control in LDM-based Talking-Head GenerationDiffusion models → Audio, speech and music
EvolveNav: Empowering LLM-Based Vision-Language Navigation via Self-Improving Embodied ReasoningLLM reasoning → Spatial and embodied reasoning +1
LongDWM: Cross-Granularity Distillation for Building a Long-Term Driving World ModelWorld models → World models for autonomous driving +1
Adaptive Destruction Processes for Diffusion SamplersDiffusion models → Diffusion foundations and sampling
G4Seg: Generation for Inexact Segmentation Refinement with Diffusion ModelsDiffusion models → Medical imaging and segmentation
LAMARL: LLM-Aided Multi-Agent Reinforcement Learning for Cooperative Policy GenerationLLM agents and tool use → Reinforcement learning for LLM agents
STORM-BORN: A Challenging Mathematical Derivations Dataset Curated via a Human-in-the-Loop Multi-Agent FrameworkLLM reasoning → Mathematical reasoning +1
Learning Abstract World Models with a Group-Structured Latent SpaceWorld models → Latent predictive world models
Beyond RLHF: A Unified Theoretical Framework of AlignmentLLM alignment / RLHF / preference optimisation → LLM preference optimization methods
FormFactory: An Interactive Benchmarking Suite for Multimodal Form-Filling AgentsLLM agents and tool use → GUI and mobile agents

All 499