HotPaper.ai — 매일 자동 큐레이션되는 AI 논문 Top 25 + 한국어 요약

1. #1VideoKR: Towards Knowledge- and Reasoning-Intensive Video Understanding

2. #2TPA-AD: A Two-Stage Pseudo Anomaly-Guided Method for Bearing Time-Series Anomaly Detection

   Xiancheng Wang, Zhibo Zhang, Ran Li, Rui Wang, Minghang Zhao

3. #3Personal AI Agent for Camera Roll VQA

4. #4Dream.exe: Can Video Generation Models Dream Executable Robot Manipulation?

5. #5Large Language Models Hack Rewards, and Society

6. #6ArcANE: Do Role-Playing Language Agents Stay in Character at the Right Time?

7. #7TIDE: Proactive Multi-Problem Discovery via Template-Guided Iteration

8. #8AdaPlanBench: Evaluating Adaptive Planning in Large Language Model Agents under World and User Constraints

9. #9Reinforcement Learning Elicits Contextual Learning of Unseen Language Translation

10. #10RobotValues: Evaluating Household Robots When Human Values Conflict

11. #11AffordanceVLA: A Vision-Language-Action Model Empowering Action Generation through Affordance-Aware Understanding

12. #12Flash-WAM: Modality-Aware Distillation for World Action Models

13. #13LoomVideo: Unifying Multimodal Inputs into Video Generation and Editing

14. #14Complexity-Balanced Diffusion Splitting

15. #15Rethinking Continual Experience Internalization for Self-Evolving LLM Agents

16. #16Unsupervised Skill Discovery for Agentic Data Analysis

17. #17Uncertainty Aware Functional Behavior Prediction and Material Fatigue Assessment for Circular Factory

18. #18Spectral Scaling Laws of Muon

19. #19Code2LoRA: Hypernetwork-Generated Adapters for Code Language Models under Software Evolution

20. #20LLMs Can Leak Training Data But Do They Want To? A Propensity-Aware Evaluation of Memorization in LLMs

21. #21The Shadow Price of Reasoning: Economic Perspective on Optimal Budget Allocation for LLMs

22. #22OPRD: On-Policy Representation Distillation

23. #23Meta-Cognitive Memory Policy Optimization for Long-Horizon LLM Agents

24. #24World-Language-Action Model for Unified World Modeling, Language Reasoning, and Action Synthesis

25. #25Imagine Before You Predict: Interleaved Latent Visual Reasoning for Video Event Prediction

인기 키워드

reinforcement-learning (225) llm (139) diffusion-models (116) llm-agents (109) vlm (98) transformer (82) llm-evaluation (77) video-generation (68) long-context (55) benchmark-evaluation (53) retrieval-augmented (50) vision-language (50) code-generation (49) flow-matching (48) long-horizon (47) text-to-image (46) large-language-models (44) mllm (44) fine-tuning (41) foundation-models (41)