Metraipolis › archive › dc_d64b5ebb36bc

Verified Learning for Compiler Optimization: An LLM-Guided Architecture with Formal Control — arXiv:2609.27214

Source: arXiv (en)
Original: https://arxiv.org/abs/2609.27214
License (as archived): CC0-1.0
Archived: 2026-10-08T11:06:24.715Z
SHA-256 of the archived text: 31cf2f39e4e09b4ff21aab43a5e39da24e6a44fbc38267ae5827ea162563c971
Stable URL of this page: https://metraipolis.com/documento/dc_d64b5ebb36bc

How to cite

Verified Learning for Compiler Optimization: An LLM-Guided Architecture with Formal Control — arXiv:2609.27214 — arXiv (en), https://arxiv.org/abs/2609.27214 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_d64b5ebb36bc (Metraipolis archive dc_d64b5ebb36bc, sha256 31cf2f39e4e09b4ff21aab43a5e39da24e6a44fbc38267ae5827ea162563c971)

BibTeX

@misc{dcd64b5ebb36bc,
  title = {Verified Learning for Compiler Optimization: An LLM-Guided Architecture with Formal Control — arXiv:2609.27214},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.27214},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_d64b5ebb36bc; sha256 31cf2f39e4e09b4ff21aab43a5e39da24e6a44fbc38267ae5827ea162563c971},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.27214. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Compiler optimizations traditionally rely on handcrafted heuristics that often fail to generalize across programs and architectures. We investigate whether large language models can participate in compiler optimization through a verification-centered systems architecture that couples generative rewriting with formal equivalence checking. Using lazification in LLVM IR as a case study, we fine-tune a code-centric LLM on transformations produced by Wyvern and embed Alive2 into a feedback loop that enforces semantic preservation for every generated rewrite. Correctness is enforced externally as a runtime control layer rather than learned implicitly. During inference, candidate transformations are symbolically validated and regenerated when necessary, ensuring accepted rewrites satisfy formal constraints. On the LLVM test suite, the fine-tuned model reproduces core optimization behaviors while applying fewer transformations overall. Although Wyvern remains faster on most benchmarks, 9.8% achieve comparable or improved runtime under the learned system, with no semantic violations observed. Verification overhead remains bounded and convergence stable. These results demonstrate that generative AI components can be safely integrated into compiler pipelines through deterministic validation and structured feedback, offering a scalable architectural pattern for trustworthy AI-driven software infrastructure.

Tema degli esploratori: calcolo e infrastrutture (retro-catalogo arXiv: annunci del 2026-09-24 in cs.SE).
Categorie arXiv: Software Engineering (cs.SE).
Pagina del lavoro: https://arxiv.org/abs/2609.27214 — 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 2026-10-08T10:13:35.107Z dalla pagina https://arxiv.org/catchup/cs.SE/2026-09-24?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).