# DMAT: Direct Action-Manifold Modeling for Efficient and Robust Bimanual Manipulation

- Source: arXiv cs.RO
- Published: 2026-11-02T05:00:00.000Z
- RoboHot: https://robohot.news/en/items/bz0agsn0wjhdqnfym4lhqm9zx
- Original: https://arxiv.org/abs/2610.41643

## Summary

The authors propose the Direct Manifold Action Transformer (DMAT), which reframes bimanual action generation as directly modeling clean, observation-conditioned joint-space action chunks from stochastic interpolants, enabling single-step inference without iterative denoising or trajectory straightening.