LATIDIA · Investigación

Caracterización de la búsqueda web por agentes conversacionales de LLM: desde decisiones y estrategias de búsqueda hasta resultados y respuestas

arXiv: 2609.19244v1Announce Type: new Resumen: Los agentes de LLM conversacional dependen cada vez más de la búsqueda web, sin embargo, el ciclo de vida de extremo a extremo de la búsqueda agéntica sigue siendo poco conocido. Presentamos el primer estudio de Web searc

La noticia

arXiv:2609.19244v1 Announce Type: new Abstract: Conversational LLM agents increasingly rely on Web search, yet the end-to-end lifecycle of agentic search remains poorly understood. We present the first study of Web search across four major conversational platforms (ChatGPT, Claude, Grok, and DeepSeek), combining real-world user interactions (invivo) with controlled experiments using the same platform's models by their APIs (invitro). We investigate the quality of agentic decisions to invoke Web search, their strategies to formulate queries, the potential domain preferences in the search results they receive, and the choices they make when transforming search results into grounded responses. We find that Web-search decisions vary substantially across platforms and models, while more frequent