| title | TILDE: Teacher-in-the-Loop Distillation from Embodied Foundation Models |
|---|---|
| authors | Yuxiang Lu, Yantong Liu, Jinghua Hou, Xueli Sun, Ruihua Han, Zhe Liu, Hengshuang Zhao |
| date | 2026-10-18 |
| venue | arXiv (preprint, not peer reviewed) |
| id | arXiv:2610.25376 |
| affiliation | The University of Hong Kong |
| upvotes | 8 (Hugging Face Daily Papers, 2026-11-01) |
| links | abs pdf hf code |
| abstract | Large-scale action pretraining equips Embodied Foundation Models (EFMs) with strong manipulation capabilities, with particular strengths in generalization and closed-loop robustness, but reproducing such pretraining requires substantial robot data and computation. To transfer these capabilities from pretrained EFMs to policies without action pretraining, we introduce TILDE (Teacher-In-the-Loop Distillation from EFMs), which synergizes on-policy distillation at student-induced states with learning from teacher-executed recovery trajectories. We further propose AutoDAgger, an autonomous data aggregation pipeline in which a robotic reward model is utilized to monitor task progress and trigger teacher intervention, thereby collecting recovery experience without human monitoring or manual control. AutoDAgger accelerates recovery-trajectory collection by more than 2times compared with the human-in-the-loop method. By unifying learning from demonstrations, on-policy teacher supervision, and teacher's corrective behavior, TILDE improves task performance by up to 4.1times over behavior cloning across three real-world tasks and supports effective cross-architecture distillation. |
cite
@misc{lu2026tilde,
title={TILDE: Teacher-in-the-Loop Distillation from Embodied Foundation Models},
author={Yuxiang Lu and Yantong Liu and Jinghua Hou and Xueli Sun and Ruihua Han and Zhe Liu and Hengshuang Zhao},
year={2026},
eprint={2610.25376},
archivePrefix={arXiv},
url={https://arxiv.org/abs/2610.25376}
}