Structured but Silent: Probing Capability Requirements in LLM Hidden States
Reliable tool use requires more than triggering a mechanism or matching a query to an API description. Before selecting a specific tool, an agent must first infer the capability requirements implied by the user query. In this paper, we investigate whether these query-side capability requirements are linearly decodable...
Kyo-Nam Choo, Mi-Ra Song, Yun-Ju Kang et al.
· 0 citations