LATIDIA · Investigación

Toolcompass: prueba de herramientas de guía, no supresión

arXiv: 2609.25678v1Tipo de anuncio: nuevo Resumen: los agentes del modelo de lenguaje grande (LLM) deben generalizar de las herramientas vistas durante la capacitación a las herramientas invisibles en la implementación. Un desafío clave es la prueba de herramientas, es decir, pruebas excesivas de

La noticia

arXiv:2609.25678v1 Announce Type: new Abstract: Large language model (LLM) agents must generalize from tools seen during training to unseen tools at deployment. A key challenge is tool trialing, i.e., excessive trials waste the interaction budget, whereas selective trials enable exploration of unfamiliar tools. Existing outcome-based post-training leaves wasteful trials unguided, while turn-level supervision may suppress necessary exploration. We introduce ToolCompass, a post-training framework that guides tool trialing by organizing tool-call representations according to shared functions. Specifically, ToolCompass models each function class as a von Mises--Fisher distribution and jointly reduces intra-function variation across domains and increases inter-function separation. This structure transfers experience from seen tools to functionally similar unseen tools, directing