ReSI: Recursive Safety Improvement toward Resistant and Resilient AI
ReSI recursively red-teams and retrains models, cutting mean X-Teaming success from 86.01% to 31.45%.
1 stories, newest first · one card per story
ReSI recursively red-teams and retrains models, cutting mean X-Teaming success from 86.01% to 31.45%.
Today 1.2K readers · 1.8K views · 1.4K reads · all time 160K views