LATIDIA · Investigación

GraphEcho: redundancia estructural y procedencia de la evidencia en agentes gráficos de LLM

arXiv:2609.17695v1 Tipo de anuncio: nuevo Resumen: Un agente de modelo de lenguaje grande (LLM) puede seguir más rutas de grafos sin adquirir más evidencia independiente. GraphEcho comprueba si los agentes confunden estos encuentros repetidos

La noticia

arXiv:2609.17695v1 Announce Type: new Abstract: A large language model (LLM) agent can follow more graph paths without acquiring more independent evidence. GraphEcho tests whether agents mistake these repeated encounters for additional corroboration. The benchmark varies path counts and evidential origins while holding evidence content fixed, and evaluates both judgments and active exploration. Controlled synthetic experiments reveal model-dependent judgment shifts, but redundant supporting paths increase the share of repeated walks across all evaluated frozen agents. Provenance-aware post-training (PAPT) reduces revisits and improves synthetic accuracy, yet covers fewer distinct sources. On scientific claims, it continues to reduce repetition while accuracy declines. These findings expose a gap between efficient exploration and effective evidence