# NTU MMLab and Shanghai AI Lab propose PhysLDM: high-fidelity physics simulation with latent diffusion

- Source: 机器之心 · 微信公众号
- Published: 2026-11-02T23:13:42.000Z
- RoboHot: https://robohot.news/en/items/lynup7r7uyq1tead9i8rmu8np
- Original: https://mp.weixin.qq.com/s?__biz=MzA3MzI4MjgzMw==&mid=2651071670&idx=3&sn=5bd161befef3452dba9ab2ee4fdac700&chksm=855e90eda6fc668a73246c940dcbcadfe56ff53782de33641573383ae623e4efd5944bfbcbf0&scene=126&sessionid=1793720143#rd

## Summary

NTU MMLab and Shanghai AI Lab propose PhysLDM, which uses a spatiotemporal VAE to compress an entire deformable-object simulation trajectory into latent space and then generates it in one pass with a diffusion model, achieving an average reconstruction error of 2.48 millimeters.