Metraipolis › archive › dc_7059617c9ebf
GraphAlignCoder: Aligning Program and Proof Graphs for Code Generation — arXiv:2608.11394
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2608.11394
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-19T05:42:47.169Z
- SHA-256 of the archived text
e9a27d397be68102b33bbadc34515d9763fc1a95ce4d474d56eac401bbcaee21- Stable URL of this page
https://metraipolis.com/documento/dc_7059617c9ebf
How to cite
GraphAlignCoder: Aligning Program and Proof Graphs for Code Generation — arXiv:2608.11394 — arXiv (en), https://arxiv.org/abs/2608.11394 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_7059617c9ebf (Metraipolis archive dc_7059617c9ebf, sha256 e9a27d397be68102b33bbadc34515d9763fc1a95ce4d474d56eac401bbcaee21)
BibTeX
@misc{dc7059617c9ebf,
title = {GraphAlignCoder: Aligning Program and Proof Graphs for Code Generation — arXiv:2608.11394},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2608.11394},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_7059617c9ebf; sha256 e9a27d397be68102b33bbadc34515d9763fc1a95ce4d474d56eac401bbcaee21},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2608.11394. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Code large language models (LLMs) can generate syntactically plausible programs that nevertheless violate hidden semantic constraints. Existing execution-feedback training methods identify whether a completed program fails, but provide limited supervision about how a correct solution should be organized. We introduce GraphAlignCoder, a training framework that transfers explicit correctness structure into code generation. GraphAlignCoder constructs an implementation graph that captures control and dependence among program regions. In parallel, a constrained Lean pipeline produces proof traces, from which we extract a formal proof-flow graph. The model first learns executable code together with graph-derived descriptions of why individual program regions are correct, and then consolidates this knowledge into code generation. GraphAlignCoder consistently outperforms the base model, code-only SFT, and CodeRL across all benchmarks. Compared with CodeRL, it increases the solved count from 38 to 50 on LiveCodeBench v6 and from 16 to 23 on BigCodeBench Hard, corresponding to relative gains of 31.6% and 43.8%, while also improving BigCodeBench Full from 359 to 363 tasks. The ablation study further shows that verification-graph injection produces the initial reasoning gain, while verification to code consolidation is essential for robust cross-benchmark transfer. Tema degli esploratori: calcolo e infrastrutture (retro-catalogo arXiv: annunci del 2026-08-13 in cs.SE). Categorie arXiv: Software Engineering (cs.SE). Pagina del lavoro: https://arxiv.org/abs/2608.11394 — 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 2026-10-19T05:31:21.752Z dalla pagina https://arxiv.org/catchup/cs.SE/2026-08-13?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).