1 paper · 1 filter
Jinyi Han, Yuanjian Xu, Ying Liao +6
Large language model (LLM) agents increasingly rely on skills, structured documents that specify when to act, which procedure to follow, and which tools are allowed. Existing evalu…