LATIDIA · Investigación

EvoBelief: modelos mundiales falsificables para agentes de LLM que se mejoran a sí mismos

arXiv: 2610.12789v1Tipo de anuncio: nuevo Resumen: Los agentes de LLM en evolución suelen mejorar su forma de actuar a través de la experiencia, la reflexión o las habilidades acumuladas. Sin embargo, incluso los agentes con capacidad procesal pueden fallar cuando su beli

La noticia

arXiv:2610.12789v1 Announce Type: new Abstract: Self-evolving LLM agents typically improve how they act through accumulated experience, reflection, or skills. Yet even procedurally capable agents can fail when their beliefs about environment dynamics are incomplete, incorrect, or outdated. We study a complementary axis of self-evolution: evolving those beliefs during deployment. Such beliefs can support multiple tasks, but incorrect ones can also propagate across them. We introduce EvoBelief, a self-evolving world model that represents environment knowledge as falsifiable hypotheses. Unexpected outcomes and decision-relevant unknowns trigger hypothesis formation. Later observations and controlled experiments provide independent validation before a rule can guide predictions or plans within its supported scope. Counterexamples narrow or retire rules