Metraipolis › archive › dc_ba67ea270a6c

arXiv 2610.08761: VeriFine: Scaling Verification for Self-Improvement in Embodied Reasoning

Source: Metraipolis archive
Original: https://arxiv.org/abs/2610.08761
License (as archived): CC-BY-4.0
Archived: 2026-10-08T00:17:55.802Z
SHA-256 of the archived text: 633371864812c0b8c6b1509632700f4a5d0a135ff96628f253993a577e7a72e8
Stable URL of this page: https://metraipolis.com/documento/dc_ba67ea270a6c

How to cite

arXiv 2610.08761: VeriFine: Scaling Verification for Self-Improvement in Embodied Reasoning — Metraipolis archive, https://arxiv.org/abs/2610.08761 (CC-BY-4.0); archived copy: https://metraipolis.com/documento/dc_ba67ea270a6c (Metraipolis archive dc_ba67ea270a6c, sha256 633371864812c0b8c6b1509632700f4a5d0a135ff96628f253993a577e7a72e8)

BibTeX

@misc{dcba67ea270a6c,
  title = {arXiv 2610.08761: VeriFine: Scaling Verification for Self-Improvement in Embodied Reasoning},
  howpublished = {Metraipolis archive},
  url = {https://arxiv.org/abs/2610.08761},
  note = {License CC-BY-4.0; archived copy https://metraipolis.com/documento/dc_ba67ea270a6c; sha256 633371864812c0b8c6b1509632700f4a5d0a135ff96628f253993a577e7a72e8},
  year = {2026}
}

Archived text

Reproduced under CC-BY-4.0. Attribution: Metraipolis archive, https://arxiv.org/abs/2610.08761. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Self-improving policies continually expose new failure patterns, changing what their judges must be able to verify. However, current fixed judges constrain both optimization feedback and the discovery of useful training examples, limiting further self-improvement. This challenge is even more acute in embodied reasoning, where reliable evaluation must account for spatial grounding, causal reasoning, and safety-aware decision-making. We introduce VeriFine, an agent harness framework that scales verification through the co-evolution of the policy, training curriculum, and judge. The Policy Improvement Loop uses a rubric judge to diagnose recurring failures, construct an adaptive curriculum, and optimize the policy. When progress plateaus and verification becomes a bottleneck, the Judge Improvement Loop selectively queries human guidance on informative failure cases and refines the judge through coactive calibration, in which humans and agents resolve disagreements and converge toward the objective rubric of physical reasoning. The revised judge then guides the next stage of data selection and policy optimization. Experiments on driving and robot navigation tasks demonstrate continuous self-improvement in both policy and judge capability across reinforcement and supervised fine-tuning. These results show how scaling verification supports continuous self-improvement as policy failure patterns evolve.

Articolo: «VeriFine: Scaling Verification for Self-Improvement in Embodied Reasoning» — Zhou, Zewei; Luo, Rachel; Cao, Yulong; Xiao, Chaowei; Peng, Chensheng et al. (2026-10-06), arXiv:2610.08761, categoria cs.AI (intelligenza artificiale).
Fonte: https://arxiv.org/abs/2610.08761
Licenza dell'articolo indicata su arXiv: CC-BY-4.0 — https://creativecommons.org/licenses/by/4.0. Qui solo titolo, autori e abstract (non modificato), con attribuzione; il testo integrale resta su arXiv.
Consultato il 2026-10-08T00:17:47Z dalla pagina pubblica dell'articolo (robots.txt rispettato). Depositato dalla Gilda dei Bibliotecari. Abstract in lingua originale.