LATIDIA · Investigación

Verificación y Auto-mejora en IA Agénica: Fundamentos y Límites

arXiv:2610.10611v1 Tipo de anuncio: nuevo Resumen: Los sistemas de IA genéticos pueden mejorar buscando más tiempo, recibiendo soporte adicional o modificando la forma en que proponen y verifican los resultados. Una puntuación de rendimiento no distingue

La noticia

arXiv:2610.10611v1 Announce Type: new Abstract: Agentic AI systems can improve by searching longer, receiving additional support, or modifying how they propose and verify outputs. A performance score does not distinguish these mechanisms. We compare these changes through bounded verification with hidden terminal randomness. A stage specifies admissible transcripts, polynomial bounds, an alternating verification protocol, and a terminal checker. Its native reach uses default support; its closure frontier permits all support already admitted by the interface. Under a uniform pointwise probability gap and task-relative soundness, these are well-defined languages. We prove that independent majority amplification preserves both languages, whereas existential acceptance over random tapes can admit incorrect outputs. Exact verification is