Metraipolis › archive › dc_92803b526248
Teaching a Tree to Recover: Online Distillation of an LLM Supervisor with Counterfactual Replay — arXiv:2610.39775
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2610.39775
- License (as archived)
- CC0-1.0
- Archived
- 2026-11-01T02:45:42.525Z
- SHA-256 of the archived text
03dd448e24eb5b20abd0eb98ebe8ee9014365a93f3eed0162a6822e13ca23257- Stable URL of this page
https://metraipolis.com/documento/dc_92803b526248
How to cite
Teaching a Tree to Recover: Online Distillation of an LLM Supervisor with Counterfactual Replay — arXiv:2610.39775 — arXiv (en), https://arxiv.org/abs/2610.39775 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_92803b526248 (Metraipolis archive dc_92803b526248, sha256 03dd448e24eb5b20abd0eb98ebe8ee9014365a93f3eed0162a6822e13ca23257)
BibTeX
@misc{dc92803b526248,
title = {Teaching a Tree to Recover: Online Distillation of an LLM Supervisor with Counterfactual Replay — arXiv:2610.39775},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2610.39775},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_92803b526248; sha256 03dd448e24eb5b20abd0eb98ebe8ee9014365a93f3eed0162a6822e13ca23257},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.39775. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Autonomous mobile robots commonly pair a global planner with a local model predictive controller (MPC), but nonconvex obstacle avoidance and a limited horizon can make the robot freeze in complex layouts. Deciding whether to replan, wait, or call for help is a supervisory decision. Recent work queries a large language model (LLM) supervisor on every anomaly, which is slow and costly. We instead distill the LLM online into a decision tree. After each training episode, counterfactual replay labels recoveries as needed, unnecessary, or harmful. These labels retire unhelpful rules. In simulation, on 20 unseen layouts where the MPC alone freezes, the tree handles at least 18 correctly across all seeds, either by reaching the goal or by calling for help when no feasible path exists. The tree matches both an always-query LLM supervisor and expert-designed rules, while issuing no LLM queries at test time. Handwritten rules cannot anticipate the long tail of situations a physical robot will encounter. Online distillation is a promising remedy: the tree handles familiar cases, and the LLM is consulted only for new ones, whose validated recoveries become new rules. Language models can thus also serve as designers of supervisors rather than only as runtime supervisors. Tema degli esploratori: statistica e scienza dei dati (elenco arXiv cs.RO, nuovi lavori del giorno). Categorie arXiv: Robotics (cs.RO). Pagina del lavoro: https://arxiv.org/abs/2610.39775 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito). Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina. Letto il 2026-11-01T02:16:58.668Z dalla pagina https://arxiv.org/list/cs.RO/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).