LAR-MoE: Latent-Aligned Routing for Mixture of Experts in Robotic Imitation Learning
Fuente:
arXiv
Salvato in:
| Autori principali: | , , , , , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2026
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866914380070780928 |
|---|---|
| author | Rodriguez, Ariel Li, Chenpan Mazza, Lorenzo Younis, Rayan Hellig, Ortrun Bodenstedt, Sebastian Wagner, Martin Speidel, Stefanie |
| author_facet | Rodriguez, Ariel Li, Chenpan Mazza, Lorenzo Younis, Rayan Hellig, Ortrun Bodenstedt, Sebastian Wagner, Martin Speidel, Stefanie |
| contents | Imitation learning enables robots to acquire manipulation skills from demonstrations, yet deploying a policy across tasks with heterogeneous dynamics remains challenging, as models tend to average over distinct behavioral modes present in the demonstrations. Mixture-of-Experts (MoE) architectures address this by activating specialized subnetworks, but requires meaningful skill decompositions for expert routing. We introduce Latent-Aligned Routing for Mixture of Experts (LAR-MoE), a two-stage framework that decouples unsupervised skill discovery from policy learning. In pre-training, we learn a joint latent representation between observations and future actions through student-teacher co-training. In a post-training stage, the expert routing is regularized to follow the structure of the learned latent space, preventing expert collapse while maintaining parameter efficiency. We evaluate LAR-MoE in simulation and on hardware. On the LIBERO benchmark, our method achieves a 95.2% average success rate with 150M parameters. On a surgical bowel grasping and retraction task, LAR-MoE matches a supervised MoE baseline without requiring any phase annotations, and transfers zero-shot to ex vivo porcine tissue. Our findings suggest that latent-aligned routing provides a principled alternative to supervised skill decomposition, enabling structured expert specialization from unlabeled demonstrations. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2603_08476 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | LAR-MoE: Latent-Aligned Routing for Mixture of Experts in Robotic Imitation Learning Rodriguez, Ariel Li, Chenpan Mazza, Lorenzo Younis, Rayan Hellig, Ortrun Bodenstedt, Sebastian Wagner, Martin Speidel, Stefanie Robotics Imitation learning enables robots to acquire manipulation skills from demonstrations, yet deploying a policy across tasks with heterogeneous dynamics remains challenging, as models tend to average over distinct behavioral modes present in the demonstrations. Mixture-of-Experts (MoE) architectures address this by activating specialized subnetworks, but requires meaningful skill decompositions for expert routing. We introduce Latent-Aligned Routing for Mixture of Experts (LAR-MoE), a two-stage framework that decouples unsupervised skill discovery from policy learning. In pre-training, we learn a joint latent representation between observations and future actions through student-teacher co-training. In a post-training stage, the expert routing is regularized to follow the structure of the learned latent space, preventing expert collapse while maintaining parameter efficiency. We evaluate LAR-MoE in simulation and on hardware. On the LIBERO benchmark, our method achieves a 95.2% average success rate with 150M parameters. On a surgical bowel grasping and retraction task, LAR-MoE matches a supervised MoE baseline without requiring any phase annotations, and transfers zero-shot to ex vivo porcine tissue. Our findings suggest that latent-aligned routing provides a principled alternative to supervised skill decomposition, enabling structured expert specialization from unlabeled demonstrations. |
| title | LAR-MoE: Latent-Aligned Routing for Mixture of Experts in Robotic Imitation Learning |
| topic | Robotics |
| url | https://arxiv.org/abs/2603.08476 |