2024/06/12 by Chloe Loughridge, Qinyi Sun, Loughridge, Chloe +17 · 16 citations
Computer Science · #Artificial Intelligence (cs.AI) #FOS: Computer and information sciences #Formal Methods in Verification #Machine Learning (cs.LG) #Programming Languages (cs.PL) #Security and Verification in Computing #Software Engineering (cs.SE) #Software Testing and Debugging Techniques
paper · pdf · doi:10.48550/arxiv.2406.08467
openalex publication_date 2024/06/12 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28
We introduce DafnyBench, the largest benchmark of its kind for training and evaluating machine learning systems for formal software verification. We test the ability of LLMs such as GPT-4 and Claude 3 to auto-generate enough hints for the Dafny formal verification engine to successfully verify over 750 programs with about 53,000 lines of code. The best model and prompting scheme achieved 68% success rate, and we quantify how this rate improves when retrying with error message feedback and how it deteriorates with the amount of required code and hints. We hope that DafnyBench will enable rapid improvements from this baseline as LLMs and verification techniques grow in quality.