HuggingFace Papers

• HuggingFace Papers Today Week Month
• 共 105 条
• 1 ChronoReplay: One Answer Interface Is Not Enough for Streaming Spatial Intelligence

  2610.34283 ▲ 399
• 2 The Trinity:Capability-Driven Recursive Self-Improvement for Multimodal Agents

  2610.36077 ▲ 384
• 3 Multiscale Policy Optimization for Long-Horizon LLM Agents

  2610.35013 ▲ 379
• 4 Trajectory2Gym: Decision-Grounded Compositional Interaction Synthesis for Tool-Use Agents

  2610.33568 ▲ 367
• 5 Trial by Trace: Benchmarking AI Agent Harnesses for Reliability and Failure Attribution

  2610.32285 ▲ 299
• 6 Reading, Writing, Sharing, and Interacting with the World through Video

  2610.31668 ▲ 295
• 7 StreamVibe: Proactive Video Agents for Evolving User Intent

  2610.34662 ▲ 171
• 8 Memory Tax: Dissecting the Net Value of Agent Memory

  2610.34073 ▲ 149
• 9 Efficient Streaming Audio-Visual Target Speaker Extraction for Real-World Acoustic Scenes

  2610.32749 ▲ 147
• 10 PDRVLA: A Vision-Language-Action Model with Physical Dynamics Representations

  2610.37381 ▲ 143
• 11 DynaFlow: Learning Free-Form Multi-Agent Orchestration

  2610.34467 ▲ 142
• 12 Function-Space Temporal Self-Distillation for Policy Optimization

  2610.19687 ▲ 140
• 13 InteractiveWorld: A Verifiable Human-Centric Environment for Building Better Interactive MLLM Assistants

  2610.33966 ▲ 126
• 14 Hy-Nomas: Towards Free-form Interactive World Generation

  2610.36729 ▲ 118
• 15 Work Agents: Systems, Capabilities, and Optimization

  2610.35845 ▲ 118
• 16 Can Test-Time Memory Evolution Outperform Test-Time Training for Scientific Discovery?

  2610.32046 ▲ 112
• 17 VibeVideoBench: Evaluating Multimodal Agents on Code-Driven Video Generation

  2610.35864 ▲ 89
• 18 HABIT-Bench: Evaluating Habit Induction from Longitudinal Weak Evidence in Agent Memory

  2610.24520 ▲ 84
• 19 Rollback: Experience Backtracking in World Models for Policy Improvement

  2610.36068 ▲ 81
• 20 UserArena: Benchmarking Interactive User Simulation for Long-Horizon Shopping Agents

  2610.30932 ▲ 63
• 21 Representation Redistribution in Sparse Autoencoders

  2610.21240 ▲ 61
• 22 Continual Learning Exam: Benchmarking Long-Horizon RSI and Experiential Learning in LLM Agents

  2610.33956 ▲ 61
• 23 HiSTEVE 1.0: Harnessing Human-Interface VLA Agents for Open-World Games

  2610.31162 ▲ 57
• 24 RynnUnify: Towards Unified Embodied Brain

  2610.32965 ▲ 55
• 25 Spatial Audio Generation and Editing with Executable Plans and Self-Distilling Transfusion

  2610.34415 ▲ 54
• 26 WorldCast: World-Consistent Character Video Generation with Large-Angle Camera Control

  2610.33135 ▲ 54
• 27 Can Discrete Tokens Serve as Semantically Lossless Representations of Visual Signals?

  2610.37662 ▲ 53
• 28 ThinkWorld: World Imagination Reinforces Chain-of-Dynamics Reasoning

  2610.37666 ▲ 52
• 29 CodeCap-Bench: Can Code Serve as a Good Video Caption?

  2610.37341 ▲ 49
• 30 RSIHub: A Composable Framework for Agent-Evolution Methods

  2610.32253 ▲ 49
• 31 Eat Your Precision: Precision Objective Mitigates Reward Hacking in Rubric-based RL

  2610.33512 ▲ 48
• 32 AV-FABLE: Audio-Visual Manipulation Forensics in Long-Form Videos

  2610.33413 ▲ 46
• 33 LociArt: A Unified Paradigm for Free-Style Local Video Stylization

  2610.36456 ▲ 44
• 34 HiP3R: Hierarchical Pose Graphs for Long-Horizon 3D Reconstruction

  2610.31635 ▲ 42
• 35 OmniShowcase: Omnidirectional Subject-Consistent Video Generation from Multi-View Images

  2610.35684 ▲ 41
• 36 Qwen-Image-PE: Learning Visual Design for Text-to-Image Prompt Enhancement

  2610.33472 ▲ 41
• 37 Principia: Test-Time Hypotheses to Principle Construction for Scientific Discovery

  2610.32195 ▲ 41
• 38 Omni-Woodpecker: Hallucination Correction via Evidence-Prior Rebalancing for Omni LLMs

  2610.27103 ▲ 40
• 39 On-Policy Self-Training: Data-Free Self-Improvement in Large Language Models

  2610.32770 ▲ 39
• 40 World-in-Context: Interactive Video World Modeling with In-Context Conditioning

  2610.33169 ▲ 36
• 41 Dual Exam: Controlled Evaluation of Multimodal Agents and World Models

  2610.31066 ▲ 36
• 42 S3AC: Stabilized, Selective, and Scalable Actor-Critic for LLM Reasoning

  2610.35518 ▲ 33
• 43 DirCast: Live Esports Directing and Commentary with a Single Streaming Video-Language Model

  2610.31734 ▲ 32
• 44 Evolving to Match Capabilities: Improving Harness Self-Evolution via Skill Adaptation

  2610.33024 ▲ 31
• 45 RSI Arena: Towards Recursive Self-Improvement with Human Feedback

  2610.33624 ▲ 30
• 46 RynnVLA-Latent: Scaling Robotic Manipulation Foundation Models with 3D-Grounded Latent Actions

  2610.34356 ▲ 28
• 47 Δτ-Bench: Pushing Agents Toward a Moving Capability

  2610.34238 ▲ 28
• 48 DiscoDemo: Discovering Efficient and Diverse Robot Demonstrations for Imitation Learning

  2610.32093 ▲ 28
• 49 ProAS: A Progress-Aware Advantage Shaping Framework for Multi-Turn Agentic Reinforcement Learning

  2610.18683 ▲ 27
• 50 Zooming without Zooming v2: From Seeing Details to Connecting Evidence for Generalizable Multimodal Perception

  2610.37037 ▲ 26
• 51 Nemotron-Labs-Diffusion-Uni: Elastic Self-Refining Unified Diffusion LLM with Cross-Paradigm On-Policy Distillation

  2610.36466 ▲ 26
• 52 Mycelium: A Meta-Study of Hallucination Detection Across Provenance, Typology, and Granularities

  2610.33608 ▲ 26
• 53 Build, Don't Play: Self-Improving LLM Research Agents Develop a Hearthstone Battlegrounds AI That Rivals Top Players

  2610.33845 ▲ 26
• 54 Trustworthy Recursive Self-Improvement in the Era of Large Models: Foundation, Risks, Evaluation and Open Challenges

  2610.35193 ▲ 26
• 55 EmbeddingGemma 2: Unified and Efficient Multimodal Representations

  2610.37364 ▲ 25
• 56 Speculative Diffusion is a Tale of Two Tradeoffs

  2610.31819 ▲ 25
• 57 Every Weight Is a Path of Shared Discrete Decisions

  2610.31931 ▲ 25
• 58 What Effort Level Should I Choose? On the Economics of Reasoning Effort in LLM Agents

  2610.34491 ▲ 24
• 59 miniTE: Diagnosing and Bridging the Execution Gap in Native NVFP4 Reinforcement Learning

  2610.23928 ▲ 24
• 60 More Memory, Better Videos? A Controlled Study of Visual Memory in Long-Video Generation

  2610.35001 ▲ 23
• 61 On the SPOT: Enhancing Speculative Decoding for Diffusion Language Models via Commit-Order Prediction

  2610.22543 ▲ 22
• 62 CausalAlign: Causal Alignment of Video Generation Models for Consistent Ego-Motion

  2610.29915 ▲ 22
• 63 RoboHarness: A Simple Harness Outperforms VLA and World Action Models

  2610.14018 ▲ 22
• 64 How Far Can Coding Agents Go in Visual Generation?

  2610.19025 ▲ 21
• 65 Sensorimotor Alignment: A Platonic Proxy for Embodied Policy Success

  2610.31670 ▲ 20
• 66 iPrefBench: A User-Centric Benchmark for Personalized Mobile Use Agents

  2610.34735 ▲ 20
• 67 Extreme Compression of Per-Layer Embeddings via Cross-Layer Latent Vector Quantization

  2610.31309 ▲ 20
• 68 TenetBench: A Clause-Anchored Russian Corpus for Evaluating Language Models against Professional Codes of Ethics

  2610.29477 ▲ 20
• 69 DraMaze: A Multi-Agent Pipeline with Future-Aware Memory for Diverse Microdrama Script Generation

  2610.32002 ▲ 20
• 70 Cross-modal Steering Vectors Exist Across Both Image and Text Generation

  2610.36269 ▲ 19
• 71 Thinking-in-Bets at Test Time Reduces Sycophantic Revisions in LLM Agents

  2610.34654 ▲ 19
• 72 EmbodiedLab: Benchmarking Embodied Agents That Act to Know

  2610.35385 ▲ 19
• 73 h1: A Taste of the Bitter Lesson in Automated Theorem Proving

  2610.35962 ▲ 18
• 74 MatAnyone-E: Sparse Memory Propagation for Efficient High-Resolution Video Matting

  2610.35724 ▲ 18
• 75 ReGE: Stronger Agents through Experience Synthesis

  2610.33861 ▲ 17
• 76 ParetoRSI-Gym: Can AI Agents Improve Language Models Across Multiple Domains?

  2610.34386 ▲ 16
• 77 AgentProofBench: Benchmarking Reusable Verifiers for Web Agents

  2610.34634 ▲ 16
• 78 SliderEdit++: Improved Slider Trajectories in Instruction-Based Image Editing

  2610.14253 ▲ 16
• 79 MMVistaReason: Toward Open-Data and Post-Training Recipes for Multimodal Reasoning

  2610.01352 ▲ 16
• 80 Building Auditable AI Text Detectors with Sparse Autoencoders

  2610.30006 ▲ 16
• 81 RealCLI: Toward a Unified Evaluation of Human-Agent Collaboration for CLI Agents

  2610.22507 ▲ 15
• 82 TAO: Tool-Aware Joint Scheduling for LLM-Based Agentic Serving

  2610.37230 ▲ 15
• 83 Can LLMs Uncover Hidden Physical Laws through Free Exploration?

  2610.37649 ▲ 15
• 84 Steering Black-Box Large Language Model Agents with Pitfall Anticipation World Model

  2610.29139 ▲ 15
• 85 FLOMSS: Generative Music Source Separation by Flow Matching

  2610.17821 ▲ 15
• 86 Echo: Coupled Action and Task Status Signal Denoising for Vision Language Action Models

  2610.31569 ▲ 15
• 87 Channel Selection for Mixture-of-Experts Decoding Needs No Calibration Data

  2610.37386 ▲ 14
• 88 Omni-InteractScope: Benchmarking Proactive and Agentic Interaction in Continuous Omnimodal Streams

  2610.35439 ▲ 14
• 89 Rethinking Video Storage for AI Training: Breaking Cross-Frame Dependencies for Efficient Random Access

  2610.33079 ▲ 14
• 90 Nested Sparsity for Scalable Pretraining and Continual Learning

  2610.31818 ▲ 14
• 91 Harness-Model Collaboration for Long-Horizon Continual Learning

  2610.34724 ▲ 13
• 92 RoboJev: An Open Vision-Language Judge for Robot Success Detection

  2610.33718 ▲ 13
• 93 Beyond Benchmarks: Can Small Open-Weight Models Get Real Coding Work Done?

  2610.34241 ▲ 13
• 94 Novel or Not? PatEx-Gym: Agentic Prior-Art Search for Novelty Assessment in Patent Examination

  2610.33503 ▲ 13
• 95 DuoLM: Fast and Effective Personalized Recommendation with Language Models

  2610.24481 ▲ 13
• 96 Joint or Per-Option? Compact Encoders for Closed-Set Decisions from 17 Million to 1 Billion Parameters

  2610.30970 ▲ 12
• 97 Kolab: Benchmarking Human-Agent Collaboration in Long-Horizon Workflows

  2610.21109 ▲ 12
• 98 LLM2Cert: LLM Synthesis of Certificates with SMT/Lean Verification

  2610.30524 ▲ 12
• 99 FocusVLA: Future-Intent Guided Dynamic View Routing for Vision-Language-Action Models

  2610.31423 ▲ 12
• 100 MultiAgent-OPD: End-to-End Training of Hierarchical Multi-Agent Systems with On-Policy Distillation

  2610.23234 ▲ 12
• 101 Generalized CycleGRPO for Fine-Grained Multimodal Understanding and Grounding

  2610.29552 ▲ 12
• 102 CRM-Agent-Bench: Benchmarking Tool-Using Agents in Stateful Enterprise Environments

  2610.32004 ▲ 11
• 103 DialFork: Going Beyond Factuality to Evaluate Procedural Hallucination in Conversational Agents

  2610.21561 ▲ 11
• 104 JockeyBench: Evaluating LLM Agents on Horse-Racing Decision-Making

  2610.19589 ▲ 11
• 105 Easy-3Dfy: Object-Level Geometric Composition for Single-Image 3D Scene Construction

  2610.32250 ▲ 11