Flexible Multitask Learning with Factorized Diffusion Policy
Fuente:
arXiv
Enregistré dans:
| Auteurs principaux: | , , , , , , |
|---|---|
| Format: | Preprint |
| Publié: |
2025
|
| Sujets: | |
| Accès en ligne: | |
| Tags: |
Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
|
| _version_ | 1866915953820827648 |
|---|---|
| author | Liu, Chaoqi Chen, Haonan Høeg, Sigmund H. Yao, Shaoxiong Li, Yunzhu Hauser, Kris Du, Yilun |
| author_facet | Liu, Chaoqi Chen, Haonan Høeg, Sigmund H. Yao, Shaoxiong Li, Yunzhu Hauser, Kris Du, Yilun |
| contents | Multitask learning poses significant challenges due to the highly multimodal and diverse nature of robot action distributions. However, effectively fitting policies to these complex task distributions is often difficult, and existing monolithic models often underfit the action distribution and lack the flexibility required for efficient adaptation. We introduce a novel modular diffusion policy framework that factorizes complex action distributions into a composition of specialized diffusion models, each capturing a distinct sub-mode of the behavior space for a more effective overall policy. In addition, this modular structure enables flexible policy adaptation to new tasks by adding or fine-tuning components, which inherently mitigates catastrophic forgetting. Empirically, across both simulation and real-world robotic manipulation settings, we illustrate how our method consistently outperforms strong modular and monolithic baselines. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2512_21898 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Flexible Multitask Learning with Factorized Diffusion Policy Liu, Chaoqi Chen, Haonan Høeg, Sigmund H. Yao, Shaoxiong Li, Yunzhu Hauser, Kris Du, Yilun Robotics Artificial Intelligence Multitask learning poses significant challenges due to the highly multimodal and diverse nature of robot action distributions. However, effectively fitting policies to these complex task distributions is often difficult, and existing monolithic models often underfit the action distribution and lack the flexibility required for efficient adaptation. We introduce a novel modular diffusion policy framework that factorizes complex action distributions into a composition of specialized diffusion models, each capturing a distinct sub-mode of the behavior space for a more effective overall policy. In addition, this modular structure enables flexible policy adaptation to new tasks by adding or fine-tuning components, which inherently mitigates catastrophic forgetting. Empirically, across both simulation and real-world robotic manipulation settings, we illustrate how our method consistently outperforms strong modular and monolithic baselines. |
| title | Flexible Multitask Learning with Factorized Diffusion Policy |
| topic | Robotics Artificial Intelligence |
| url | https://arxiv.org/abs/2512.21898 |