SecProbe: Adaptive Evaluation of Coding Agents on Cybersecurity Vulnerabilities
SecProbe adaptively tests coding agents on synthesized vulnerability repairs; frontier models peak at 28.33% success.
1 stories, newest first · one card per story
SecProbe adaptively tests coding agents on synthesized vulnerability repairs; frontier models peak at 28.33% success.
Today 1.1K readers · 1.2K views · 108 reads · all time 211K views