Metraipolis › archive › dc_346f1ff22787

HiRE: Hindsight Reward Editing for Policy Finetuning — arXiv:2609.27068

Source
arXiv (en)
Original
https://arxiv.org/abs/2609.27068
License (as archived)
CC0-1.0
Archived
2026-10-08T12:51:13.312Z
SHA-256 of the archived text
7d3c1ce7a34f1ecc2662745d7b8cf5a82481a543e970819da3c2559e85624cfb
Stable URL of this page
https://metraipolis.com/documento/dc_346f1ff22787

How to cite

HiRE: Hindsight Reward Editing for Policy Finetuning — arXiv:2609.27068 — arXiv (en), https://arxiv.org/abs/2609.27068 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_346f1ff22787 (Metraipolis archive dc_346f1ff22787, sha256 7d3c1ce7a34f1ecc2662745d7b8cf5a82481a543e970819da3c2559e85624cfb)
BibTeX
@misc{dc346f1ff22787,
  title = {HiRE: Hindsight Reward Editing for Policy Finetuning — arXiv:2609.27068},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.27068},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_346f1ff22787; sha256 7d3c1ce7a34f1ecc2662745d7b8cf5a82481a543e970819da3c2559e85624cfb},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.27068. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Pre-trained robot policies always require finetuning to adapt to specific environments. Reinforcement Learning (RL) offers high performance potential because it improves action optimality rather than simply mimicking data. However, such potential depends heavily on reward quality. Sparse rewards lack process feedback, human-designed rewards are costly and biased, and semantic rewards from foundation representations are often not control-centric. We propose Hindsight Reward Editing (HiRE), a training-free framework to break this reward bottleneck. HiRE bridges the broad knowledge of foundation representation models with physical control awareness, by contrasting successful and failed trajectories in hindsight. It calibrates foundation representation models by identifying "trap states" that are predicted as high-rewarding states yet eventually result in failure, and vice versa. HiRE explicitly penalizes these traps while boosting rewards for critical successful states. This approach can be flexibly compatible with any foundation representations and RL algorithms. Experiments show that HiRE consistently outperforms other reward recipes by delivering dense, control-aware feedback that prevents value function collapse and reward hacking, thereby achieving superior sample efficiency, stable policy updates, and higher performance ceilings, e.g., at least 3x performance of the base policies. Qualitative results are at this https URL .

Tema degli esploratori: statistica e scienza dei dati (retro-catalogo arXiv: annunci del 2026-09-24 in cs.RO).
Categorie arXiv: Robotics (cs.RO).
Pagina del lavoro: https://arxiv.org/abs/2609.27068 — 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 2026-10-08T12:21:19.426Z dalla pagina https://arxiv.org/catchup/cs.RO/2026-09-24?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).