Metraipolis › archive › dc_75bf344dea01
Chain of Faith: Policy-Internal Rewards for Reasoning Reliance — arXiv:2611.02672
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2611.02672
- License (as archived)
- CC0-1.0
- Archived
- 2026-11-03T11:03:18.910Z
- SHA-256 of the archived text
4e5cb3aa31b7ae1ed85189fddac8bb7eaf8f1ec7c9fc590cb10242c1a716868a- Stable URL of this page
https://metraipolis.com/documento/dc_75bf344dea01
How to cite
Chain of Faith: Policy-Internal Rewards for Reasoning Reliance — arXiv:2611.02672 — arXiv (en), https://arxiv.org/abs/2611.02672 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_75bf344dea01 (Metraipolis archive dc_75bf344dea01, sha256 4e5cb3aa31b7ae1ed85189fddac8bb7eaf8f1ec7c9fc590cb10242c1a716868a)
BibTeX
@misc{dc75bf344dea01,
title = {Chain of Faith: Policy-Internal Rewards for Reasoning Reliance — arXiv:2611.02672},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2611.02672},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_75bf344dea01; sha256 4e5cb3aa31b7ae1ed85189fddac8bb7eaf8f1ec7c9fc590cb10242c1a716868a},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2611.02672. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Chain-of-thought (CoT) explanations can sound plausible while having little influence on a model's final answer. We study four rewards that use the policy's own signals to encourage reliance on generated reasoning: gradient sensitivity, confidence gaps, progressive confidence gains, and reasoning--prompt swaps. We add each reward to group relative policy optimization (GRPO) and evaluate two small instruction-tuned models on GSM8K and StrategyQA using reasoning edits, paraphrases, and prompt hints. Several configurations increase answer reliance to reasoning edits, while effects on accuracy and paraphrase robustness vary. We also observe answer copying at a high reward weight and high reliance in decomposition settings with poor task accuracy. The study compares practical objectives for reasoning reliance, confirms the trade-off between reliance and performance, and proposes four lightweight rewards that can improve reasoning reliance. Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.AI, nuovi lavori del giorno). Categorie arXiv: Artificial Intelligence (cs.AI) ; Machine Learning (cs.LG). Pagina del lavoro: https://arxiv.org/abs/2611.02672 — 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 2026-11-03T10:59:45.840Z dalla pagina https://arxiv.org/list/cs.AI/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).