LLM-generated ACSL contracts for numerical libraries
A new approach uses large language models (LLMs) to automatically generate a small, human-certifiable Automated C Specification Language (ACSL) contract from function documentation. This combined with a deterministic toolchain that supports a restricted ACSL profile and generates a driver using the CIVL verifier checks the implementation against the contract and an existing reference model. The…
Key points
- LLM-generated ACSL contracts for numerical libraries
- Demonstrated on three PETSc functions: MatAXPY, MatAYPX, and non-compressing mode of MatFilter
- Discovered a previously undiscovered bug in MatAYPX since 1997
The headline, key points and digest above were generated by Digest AI's editorial model from the linked sources. Automated summaries can contain errors: the sources are the record. Spotted a mistake? Tell us. Published by Martin K., who runs Digest AI and handles corrections.
More in Research
All →- BioDyad integrates biomedical discovery with ML program search · 1 src
- Researchers propose MedCode to boost LLMs’ medical calculation accuracy by 20–30% · 1 src
- Researchers release open protocol for distributed AI textbook formalization · 1 src
- Researchers propose NashEval for context-dependent AI agent evaluation · 1 src
- Researchers introduce countermem to improve AI agent memory with counterfactual checks · 1 src
Comments
via GitHub Discussions