Abstracting Robot Manipulation Skills via Mixture-of-Experts Diffusion Policies, Ce Hao★, Xuanran Zhai★, Yaohua Liu, Harold Soh★, International Conference on Learning Representations (ICLR)
Links:
Diffusion-based policies have recently shown strong results in robot manipulation, but their extension to multi-task scenarios is hindered by the high cost of scaling model size and demonstrations. We introduce Skill Mixture-of-Experts Policy (SMP), a diffusion-based mixture-of-experts policy that learns a compact orthogonal skill basis and uses sticky routing to compose actions from a small, task-relevant subset of experts at each step.
A variational training objective supports this design, and adaptive expert activation at inference yields fast sampling without oversized backbones. We validate SMP in simulation and on a real dual-arm platform with multi-task learning and transfer learning tasks, where SMP achieves higher success rates and markedly lower inference cost than large diffusion baselines. These results indicate a practical path toward scalable, transferable multi-task manipulation: learn reusable skills once, activate only what is needed, and adapt quickly when tasks change.
Resources
You can find our paper here.
Citation
Please consider citing our paper if you build upon our results and ideas.
Ce Hao★, Xuanran Zhai★, Yaohua Liu, Harold Soh★, “Abstracting Robot Manipulation Skills via Mixture-of-Experts Diffusion Policies”, International Conference on Learning Representations (ICLR)
@inproceedings{hao2026abstracting, title={Abstracting Robot Manipulation Skills via Mixture-of-Experts Diffusion Policies}, author={Hao, Ce and Zhai, Xuanran and Liu, Yaohua and Soh, Harold}, booktitle={International Conference on Learning Representations (ICLR)}, year={2026} }
Contact
If you have questions or comments, please contact Ce Hao or Harold.
