Work

Data Race Detection through Vibe Translation

Year: 2025
Type: conference-paper
Abstract: We propose a data race detection approach for code written in a source programming language, by means of AI-agent translation to a target language, followed by conventional tool-based detection in the target language. We evaluate this translate-then-check approach by translating the C/Fortran+OpenMP programs in DataRaceBench to the Go programming language, and using the Go data race detector to check for races. The translation is controlled through natural language prompts, similar to approaches popularized as vibe coding. Translate-then-check achieves 92.8% accuracy and 9 false negatives for the C programs in DataRaceBench, compared to 89.9% accuracy and 17 false negatives for Clang with ThreadSanitizer and Archer applied to the original C programs. We discuss the approach and its overall accuracy, and show examples where translate-then-check leads to false negatives or positives due to limitations of the Go data race checker or, in some cases, limitations of the translation.
Authors: Jan Hueckelheim, Vimarsh Sathia, Siyuan Brant Qian
Language: English
Subfield: Software
Open Access status: hybrid
OpenAlex ID: W4416004134
FWCI:
Cites: 19
Cited by:
Related to: 10