ParetoRSI-Gym evaluates multi-domain recursive self-improvement
Claude Code with Opus-5 raised equal-weighted generalization scores 9.2% relative, or 2.3 absolute points, over the starting models.
1 stories, newest first · one card per story
Claude Code with Opus-5 raised equal-weighted generalization scores 9.2% relative, or 2.3 absolute points, over the starting models.
Today 1.4K readers · 2.3K views · 1.8K reads · all time 160K views