Metraipolis › archive › dc_226ff133faaa

ChunkTrust: Adapting Execution Horizons for Robot Policies with Action-Expert Evidence — arXiv:2609.39754

Source: arXiv (en)
Original: https://arxiv.org/abs/2609.39754
License (as archived): CC0-1.0
Archived: 2026-10-07T15:12:11.197Z
SHA-256 of the archived text: 8464772965f10b8f706d7008fcf04680df3d74af2d2ba82571640b6a8175ebe8
Stable URL of this page: https://metraipolis.com/documento/dc_226ff133faaa

How to cite

ChunkTrust: Adapting Execution Horizons for Robot Policies with Action-Expert Evidence — arXiv:2609.39754 — arXiv (en), https://arxiv.org/abs/2609.39754 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_226ff133faaa (Metraipolis archive dc_226ff133faaa, sha256 8464772965f10b8f706d7008fcf04680df3d74af2d2ba82571640b6a8175ebe8)

BibTeX

@misc{dc226ff133faaa,
  title = {ChunkTrust: Adapting Execution Horizons for Robot Policies with Action-Expert Evidence — arXiv:2609.39754},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.39754},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_226ff133faaa; sha256 8464772965f10b8f706d7008fcf04680df3d74af2d2ba82571640b6a8175ebe8},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.39754. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Robot foundation policies predict action chunks, but how many actions to execute before replanning depends on the current task phase. We introduce ChunkTrust, which treats the execution horizon as a latent variable inferred from action-expert evidence rather than a fixed hyperparameter. Its training-free Action-aware Horizon Selector (AHS) combines intra-chunk spectral stability of generation traces with inter-chunk continuity between executed history and predicted actions. An online Beta posterior with kernel forgetting tracks horizon preferences across replans. A lightweight Query-based Horizon Adapter (QHA) optionally learns a context-conditioned dense prior from complementary evidence, fused with current evidence and episode-local Beta memory while the base policy remains frozen. Across RoboTwin2.0 and RoboCasa GR1 Tabletop, AHS improves overall task-averaged success for each evaluated base-policy configuration, including gains of +6.80 percentage points on $\pi_{0.5}$ over all 50 RoboTwin2.0 tasks and +9.67 percentage points on Qwen3GR00T in RoboCasa. AHS+QHA raises the gain over Base to +9.44 percentage points on the eight-task $\pi_{0.5}$ evaluation. On four real-world household tasks, AHS improves the equal-task mean normalized process score from 50.4% to 57.5%. Ablations examine the contributions of both evidence terms, temporal memory, and the learned prior. Project page is this https URL

Tema degli esploratori: statistica e scienza dei dati (retro-catalogo arXiv: annunci del 2026-10-01 in cs.RO).
Categorie arXiv: Robotics (cs.RO).
Pagina del lavoro: https://arxiv.org/abs/2609.39754 — 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 2026-10-07T14:44:47.466Z dalla pagina https://arxiv.org/catchup/cs.RO/2026-10-01?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).