← all datasets

PatchEval

Emerging
2papers using it
88HF downloads
6HF likes
2026first seen

πŸ‘‹ Overview PatchEval is a benchmark designed to systematically evaluate LLMs and Agents in the task of automated vulnerability repair. It includes 1,000 vulnerabilities sourced from CVEs reported between 2015 and 2025, covering 65 CWE categories across Go, JavaScript, and Python. A subset of 230 CVEs is paired with Do

Papers using PatchEval (2)

PatchEval β€” datasets β€” ai-agents