Inference-Time Enhancement of Generative Robot Policies via Predictive World Modeling
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_ | 1866917333254012928 |
|---|---|
| author | Qi, Han Yin, Haocheng Zhu, Aris Du, Yilun Yang, Heng |
| author_facet | Qi, Han Yin, Haocheng Zhu, Aris Du, Yilun Yang, Heng |
| contents | We present Generative Predictive Control (GPC), an inference-time method for improving pretrained behavior-cloning policies without retraining. GPC augments a frozen diffusion policy at deployment with an action-conditioned world model trained on expert demonstrations and random exploration rollouts. The world model predicts the consequences of action proposals generated by the diffusion policy and enables lightweight online planning that ranks and refines these proposals through model-based look-ahead. By combining a generative prior with predictive foresight, GPC enables test-time adaptation while keeping the original policy fixed. Across diverse robotic manipulation tasks, including state- and vision-based settings in both simulation and real-world experiments, GPC consistently outperforms standard behavior cloning and compares favorably with other inference-time adaptation baselines. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2502_00622 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Inference-Time Enhancement of Generative Robot Policies via Predictive World Modeling Qi, Han Yin, Haocheng Zhu, Aris Du, Yilun Yang, Heng Robotics Computer Vision and Pattern Recognition Machine Learning We present Generative Predictive Control (GPC), an inference-time method for improving pretrained behavior-cloning policies without retraining. GPC augments a frozen diffusion policy at deployment with an action-conditioned world model trained on expert demonstrations and random exploration rollouts. The world model predicts the consequences of action proposals generated by the diffusion policy and enables lightweight online planning that ranks and refines these proposals through model-based look-ahead. By combining a generative prior with predictive foresight, GPC enables test-time adaptation while keeping the original policy fixed. Across diverse robotic manipulation tasks, including state- and vision-based settings in both simulation and real-world experiments, GPC consistently outperforms standard behavior cloning and compares favorably with other inference-time adaptation baselines. |
| title | Inference-Time Enhancement of Generative Robot Policies via Predictive World Modeling |
| topic | Robotics Computer Vision and Pattern Recognition Machine Learning |
| url | https://arxiv.org/abs/2502.00622 |