LATIDIA · Investigación

Un sistema LLM multiagente para la sintonización automatizada de modelos de redes neuronales especializadas

arXiv:2610.13663v1 Tipo de anuncio: nuevo Resumen: Muchos enfoques de aprendizaje automático son de dos niveles: un experto selecciona una clase de métodos, dentro de esta clase la elección de los componentes clave es crítica y generalmente manual. Esta pa

La noticia

arXiv:2610.13663v1 Announce Type: new Abstract: Many machine learning approaches are two-level: an expert selects a class of methods, within this class the choice of key components is critical and usually manual. This paper addresses the automatic tuning of these components and proposes a framework transferable across models of different nature. The multi-agent large language model (LLM) framework has three roles---analyst, coder, and recorder~---with an experiment log and a sandboxed execution environment. Unlike systems that optimize the entire ML script or search architectures in a discrete space, the framework operates over an open but semantically narrow space: the agent generates code only for the specified components, leaving the overall network structure