Metraipolis › archive › dc_755fa5b91a35

Learning to Prove, Not Just to Answer: Reinforcement Learning from Formal Verification for Natural-Language Logical Reasoning — arXiv:2609.37203

Source
arXiv (en)
Original
https://arxiv.org/abs/2609.37203
License (as archived)
CC0-1.0
Archived
2026-10-07T16:23:38.043Z
SHA-256 of the archived text
50bb0f967b2e230b47176b0c94769910b50e9ad22bd0c5bc0417c18e458a6aab
Stable URL of this page
https://metraipolis.com/documento/dc_755fa5b91a35

How to cite

Learning to Prove, Not Just to Answer: Reinforcement Learning from Formal Verification for Natural-Language Logical Reasoning — arXiv:2609.37203 — arXiv (en), https://arxiv.org/abs/2609.37203 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_755fa5b91a35 (Metraipolis archive dc_755fa5b91a35, sha256 50bb0f967b2e230b47176b0c94769910b50e9ad22bd0c5bc0417c18e458a6aab)
BibTeX
@misc{dc755fa5b91a35,
  title = {Learning to Prove, Not Just to Answer: Reinforcement Learning from Formal Verification for Natural-Language Logical Reasoning — arXiv:2609.37203},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.37203},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_755fa5b91a35; sha256 50bb0f967b2e230b47176b0c94769910b50e9ad22bd0c5bc0417c18e458a6aab},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.37203. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Large language models (LLMs) are increasingly deployed for natural-language logical reasoning, where the final answer is easy to check but the proof behind it is not. In natural-language logical reasoning, an intermediate conclusion should follow from its premises, and the resulting derivation should support the final answer. Existing methods lack machine-checkable verification of intermediate conclusions and answer-supporting proof dependencies, so they may assign credit to invalid or answer-irrelevant steps. We propose Proof-R1, an RL framework from formal verification that trains LLMs to construct verifiable proofs for natural-language logical reasoning. Proof-R1 admits a generated conclusion into the verified proof state only when the corresponding reasoning action satisfies the proof obligations through UNSAT-based machine-checkable formal verification. Proof-R1 also recovers the answer-supporting dependency closure to trace the proof structure of the final answer and align outcome credit with the proof dependencies. Experiments demonstrate that Proof-R1 improves answer accuracy across three logical reasoning benchmarks and four backbone models and outperforms training-free agents and training-based methods in terms of reasoning-process verifiability.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-30 in cs.AI).
Categorie arXiv: Artificial Intelligence (cs.AI).
Pagina del lavoro: https://arxiv.org/abs/2609.37203 — 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 2026-10-07T15:54:53.601Z dalla pagina https://arxiv.org/catchup/cs.AI/2026-09-30?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).