PhysMoDPO: Physically-Plausible Humanoid Motion with Preference Optimization

Documenti analoghi

• InterPose: Learning to Generate Human-Object Interactions from Large-Scale Web Videos
  di: Zhang, Yangsong, et al.
  Pubblicazione: (2025)
• Learning feasible transitions for efficient contact planning
  di: Akizhanov, Rikhat, et al.
  Pubblicazione: (2024)
• Selective Progress-Aware Querying for Human-in-the-Loop Reinforcement Learning
  di: Muraleedharan, Anujith, et al.
  Pubblicazione: (2025)
• U-LAG: Uncertainty-Aware, Lag-Adaptive Goal Retargeting for Robotic Manipulation
  di: H, Anamika J, et al.
  Pubblicazione: (2025)
• LikePhys: Evaluating Intuitive Physics Understanding in Video Diffusion Models via Likelihood Preference
  di: Yuan, Jianhao, et al.
  Pubblicazione: (2025)