Machine-Generated, Machine-Checked Proofs for a Verified Compiler (Experience Report)
Fuente:
arXiv
Saved in:
| Main Author: | |
|---|---|
| Format: | Preprint |
| Published: |
2026
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866914344454848512 |
|---|---|
| author | Paraskevopoulou, Zoe |
| author_facet | Paraskevopoulou, Zoe |
| contents | We report on using an agentic coding assistant (Claude Code, powered by Claude Opus 4.6) to mechanize a substantial Rocq correctness proof from scratch, with human guidance but without human proof writing. The proof establishes semantic preservation for the administrative normal form (ANF) transformation in the CertiCoq verified compiler for Rocq. The closely related continuation-passing style (CPS) transformation in CertiCoq was previously proved correct by human experts over several months. We use this proof as a template and instruct the LLM to adapt the proof technique to the ANF setting, which differs in important technical ways. The resulting ANF proof comprises approximately 7,800 lines of Rocq (larger than the 5,300-line CPS proof) and was developed in approximately 96 hours. We describe the proof technique and report on the experience of developing it with an LLM, discussing both the strengths and limitations of the approach and its implications for verified compiler construction. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2602_20082 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Machine-Generated, Machine-Checked Proofs for a Verified Compiler (Experience Report) Paraskevopoulou, Zoe Programming Languages We report on using an agentic coding assistant (Claude Code, powered by Claude Opus 4.6) to mechanize a substantial Rocq correctness proof from scratch, with human guidance but without human proof writing. The proof establishes semantic preservation for the administrative normal form (ANF) transformation in the CertiCoq verified compiler for Rocq. The closely related continuation-passing style (CPS) transformation in CertiCoq was previously proved correct by human experts over several months. We use this proof as a template and instruct the LLM to adapt the proof technique to the ANF setting, which differs in important technical ways. The resulting ANF proof comprises approximately 7,800 lines of Rocq (larger than the 5,300-line CPS proof) and was developed in approximately 96 hours. We describe the proof technique and report on the experience of developing it with an LLM, discussing both the strengths and limitations of the approach and its implications for verified compiler construction. |
| title | Machine-Generated, Machine-Checked Proofs for a Verified Compiler (Experience Report) |
| topic | Programming Languages |
| url | https://arxiv.org/abs/2602.20082 |