Metraipolis › archive › dc_50db6337f099

Compiler-Guided Adaptive Proof Search with Cross-Model Synergy on Context-Dependent Theorem Proving — arXiv:2608.18084

Source
arXiv (en)
Original
https://arxiv.org/abs/2608.18084
License (as archived)
CC0-1.0
Archived
2026-10-18T14:10:07.431Z
SHA-256 of the archived text
55007d749afcdb0e4a31d04c3c1910fcde4db41dd5a59cb2fd6e33bbad8e1b60
Stable URL of this page
https://metraipolis.com/documento/dc_50db6337f099

How to cite

Compiler-Guided Adaptive Proof Search with Cross-Model Synergy on Context-Dependent Theorem Proving — arXiv:2608.18084 — arXiv (en), https://arxiv.org/abs/2608.18084 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_50db6337f099 (Metraipolis archive dc_50db6337f099, sha256 55007d749afcdb0e4a31d04c3c1910fcde4db41dd5a59cb2fd6e33bbad8e1b60)
BibTeX
@misc{dc50db6337f099,
  title = {Compiler-Guided Adaptive Proof Search with Cross-Model Synergy on Context-Dependent Theorem Proving — arXiv:2608.18084},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2608.18084},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_50db6337f099; sha256 55007d749afcdb0e4a31d04c3c1910fcde4db41dd5a59cb2fd6e33bbad8e1b60},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2608.18084. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Theorem proving in real-world Lean 4 projects is challenging because proofs often depend on project-specific context. While iterative refinement can use compiler errors to repair failed proofs, reusing failed attempts requires careful search control: some proofs provide better starting points than others, and later revisions may degrade a partially correct proof. We propose a compiler-guided proof search framework that balances exploration and exploitation. It explores diverse starting points through dual-model generation and stagnation-triggered resampling, while exploiting promising proof states through current-best refinement guided by compiler-grounded pairwise comparison. Experiments on seven real-world Lean 4 projects from miniCTX-v2 show that our method achieves a better effectiveness--efficiency tradeoff than pass@k baselines. Within the pass@32 budget, our method improves average pass rate by 12.8 percentage points while reducing LLM calls by 21.9%.

Tema degli esploratori: calcolo e infrastrutture (retro-catalogo arXiv: annunci del 2026-08-20 in cs.PL).
Categorie arXiv: Computation and Language (cs.CL) ; Programming Languages (cs.PL).
Pagina del lavoro: https://arxiv.org/abs/2608.18084 — 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 2026-10-18T14:08:50.614Z dalla pagina https://arxiv.org/catchup/cs.PL/2026-08-20?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).