HF Daily Papers · 2026-10-22

HF Daily Papers 精选（15 篇，按社区票数排序）

  • [181▲] Matrix-Game 4.0: Harnessing Video Models for Real-Time Playable Game Worlds — https://arxiv.org/abs/2610.27999
  • [177▲] Ropedia W0: Scaling Loco-Manipulation World Models across Embodiments with Human Experience — https://arxiv.org/abs/2610.26855
  • [147▲] PAR: Progress-Aligned Recurrence for Length Generalization in State Tracking — https://arxiv.org/abs/2610.27628
  • [92▲] S-Agent 2.0: Spatial Reasoning with a General Coding Agent — https://arxiv.org/abs/2610.28354
  • [59▲] CoEvo-RSI: When Harnesses Evolve, Which Models Should Follow? — https://arxiv.org/abs/2610.25078
  • [41▲] Learning the 4D World through Point Trajectory Forecasting — https://arxiv.org/abs/2610.28594
  • [21▲] Joints, Camera, Action! Co-Generating 3D Motion and Video for Arbitrary Rigs — https://arxiv.org/abs/2610.28632
  • [21▲] SpecKCast: Spectral Kalman Forecasting for Diffusion Acceleration — https://arxiv.org/abs/2610.26837
  • [20▲] LLM2R: From Large Language Models to 3D Reassemblers — https://arxiv.org/abs/2610.18604
  • [19▲] Specter: Fast MoE Inference via Quantization-Enabled Multi-Step Speculative Offloading — https://arxiv.org/abs/2511.14102
  • [19▲] Breaking Diffusion: Parallel Reasoning with Draft Refinement — https://arxiv.org/abs/2610.26542
  • [18▲] Evolving Agent Interfaces — https://arxiv.org/abs/2610.28601
  • [17▲] Epi: Towards a Persistent, Self-Improving Agent for Autonomous Research — https://arxiv.org/abs/2610.24944
  • [17▲] EvaOmni: An Evidence-Grounded Benchmark for Audio-Visual Reasoning — https://arxiv.org/abs/2610.28444
  • [16▲] Painting Change Map as guidance for generating painting processes — https://arxiv.org/abs/2610.25369