Author
Daishiro Hirashima
Recent research
- AI & ComputingOpen access
Reproduction artefacts for the paper "When Is an LLM Code Audit Done? Estimating Audit Completeness with Capture-Recapture over Diverse LLM Lenses" (submitted to the Journal of Systems and Software). VSAT runs several deliberately diverse LLM audit lenses over one codebase, treat...
- AI & ComputingOpen access
Reproduction artefacts for the paper "When Is an LLM Code Audit Done? Estimating Audit Completeness with Capture-Recapture over Diverse LLM Lenses" (submitted to the Journal of Systems and Software). VSAT runs several deliberately diverse LLM audit lenses over one codebase, treat...
- AI & ComputingOpen access
VSAT (Vulnerability Saturation Auditing) makes the completeness of an LLM security audit a measurable, calibrated quantity. It runs several deliberately diverse LLM audit "lenses" over the same codebase and treats each lens as a capture occasion, so the overlap structure yields (...
- AI & ComputingOpen access
Tool-using LLM coding agents can reverse-generate specifications from source, but the quality of those specifications is hard to assess when no reference specification exists. We present a ground-truth-free, deterministic-first framework that measures three layers - inventory cov...
- AI & ComputingOpen access
Tool-using LLM coding agents can reverse-generate specifications from source, but the quality of those specifications is hard to assess when no reference specification exists. We present a ground-truth-free, deterministic-first framework that measures three layers - inventory cov...