Metraipolis › archive › dc_8b91e77db558

Beyond Reconstruction: What Matters in Action Tokenization for Robot Policies? — arXiv:2610.09170

Source
arXiv (en)
Original
https://arxiv.org/abs/2610.09170
License (as archived)
CC0-1.0
Archived
2026-10-08T12:48:51.945Z
SHA-256 of the archived text
0590a1f3c35b83ec8648f77e0ffd556adbbc7e25fd65312def4ad1ec77901d6b
Stable URL of this page
https://metraipolis.com/documento/dc_8b91e77db558

How to cite

Beyond Reconstruction: What Matters in Action Tokenization for Robot Policies? — arXiv:2610.09170 — arXiv (en), https://arxiv.org/abs/2610.09170 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_8b91e77db558 (Metraipolis archive dc_8b91e77db558, sha256 0590a1f3c35b83ec8648f77e0ffd556adbbc7e25fd65312def4ad1ec77901d6b)
BibTeX
@misc{dc8b91e77db558,
  title = {Beyond Reconstruction: What Matters in Action Tokenization for Robot Policies? — arXiv:2610.09170},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2610.09170},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_8b91e77db558; sha256 0590a1f3c35b83ec8648f77e0ffd556adbbc7e25fd65312def4ad1ec77901d6b},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.09170. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Autoregressive action-token policies such as vision-language-action models require action tokenizers to translate discrete token sequences into precise control actions in continuous space. Many action tokenizers learn the mapping between tokens and actions via a reconstruction objective. However, as we show through extensive analysis, sufficiently accurate action reconstruction is only one part of what makes a downstream robot policy successful. It is also critical that the policy is able to predict the right tokens for new observations, and that unseen policy token predictions still decode into reasonable actions. These properties are downstream of tokenizer training and are not directly incentivized by a reconstruction objective alone. In this work, we introduce Predictable and Robust Action Tokenization (ProAct), a tokenizer training method that strategically augments reconstruction with the goal of improving downstream predictability and robustness. ProAct is policy-agnostic and uses only action datasets for training. Across the Robomimic, LIBERO, and RoboTwin benchmarks and a diverse set of tokenizer architectures, ProAct improves rollout success by an average of 11.3 percentage points. These improvements also translate to vision-language-action policies and real-world robotic manipulation, yielding average gains of 21.8 and 36.7 percentage points, respectively. These results suggest that effective action tokenization should be designed as a policy interface that balances fidelity, predictability, and robustness, rather than as a reconstruction problem alone.

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/2610.09170 — 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 2026-10-08T12:18:30.494Z dalla pagina https://arxiv.org/list/cs.RO/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).