Metraipolis › archive › dc_8b32a67d780a
Beyond Validation Loss: Coverage and Reliability Dynamics in Robot Policy Pretraining — arXiv:2611.01565
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2611.01565
- License (as archived)
- CC0-1.0
- Archived
- 2026-11-03T11:21:08.776Z
- SHA-256 of the archived text
f289a4b2a757a9553391c3ec4a10150651476361c81daae1921d994421bc1511- Stable URL of this page
https://metraipolis.com/documento/dc_8b32a67d780a
How to cite
Beyond Validation Loss: Coverage and Reliability Dynamics in Robot Policy Pretraining — arXiv:2611.01565 — arXiv (en), https://arxiv.org/abs/2611.01565 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_8b32a67d780a (Metraipolis archive dc_8b32a67d780a, sha256 f289a4b2a757a9553391c3ec4a10150651476361c81daae1921d994421bc1511)
BibTeX
@misc{dc8b32a67d780a,
title = {Beyond Validation Loss: Coverage and Reliability Dynamics in Robot Policy Pretraining — arXiv:2611.01565},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2611.01565},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_8b32a67d780a; sha256 f289a4b2a757a9553391c3ec4a10150651476361c81daae1921d994421bc1511},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2611.01565. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
In modern robot learning, generative robot policies are first pretrained through supervised learning from demonstrations, then refined through reinforcement learning, supervised finetuning, or other forms of posttraining. Yet conventional pretraining metrics, such as validation loss and average task success, provide an incomplete account of which policies support both effective direct deployment and subsequent posttraining. In this work, we study closed-loop pass@k, which evaluates the probability that k rollouts from the same initial condition attain at least one success. Thus, pass@1 measures mean task success; large-k pass@k measures whether success remains reachable under repeated sampling. Across a wide range of posttraining recipes, we show that large-k pass@k reveals posttraining differences not captured by pass@1. In particular, checkpoints with similar pass@1 but different large-k pass@k have different posttraining performance. Moreover, we show pass@k peaks at different checkpoints during supervised pretraining, revealing a coverage-reliability tradeoff over the course of training. Based on these observations, we propose two practical interventions: (1) a purely offline checkpoint-selection criterion that balances coverage and reliability without closed-loop evaluation, and (2) a generative data augmentation procedure that expands supervision beyond the finite demonstration set, preserving large-k pass@k while maintaining or improving pass@1 and raising the observed coverage-reliability Pareto frontier. Together, our results motivate evaluation, checkpoint selection, and training procedures that explicitly account for posttraining. Blog post, visualizations, and code are available at this https URL . Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.LG, nuovi lavori del giorno). Categorie arXiv: Machine Learning (cs.LG) ; Robotics (cs.RO). Pagina del lavoro: https://arxiv.org/abs/2611.01565 — 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 2026-11-03T11:02:37.540Z dalla pagina https://arxiv.org/list/cs.LG/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).