AI Self-Generated Skills Are Useless
SkillsBench proves AI agents cannot author useful skills for themselves. Across 7,308 trajectories, self-generated skills showed zero benefit while human-curated skills improved performance by 16.2pp.
Tags
2 posts
SkillsBench proves AI agents cannot author useful skills for themselves. Across 7,308 trajectories, self-generated skills showed zero benefit while human-curated skills improved performance by 16.2pp.
MIT's SOAR framework enables LLMs to self-generate training curricula, solving the learning plateau problem in reinforcement learning.