LATIDIA · Investigación

Los agentes autónomos de IA se desvían sistemáticamente de las preferencias humanas en la formalización de pruebas

arXiv: 2611.06995v1Anuncio Tipo: nuevo Resumen: Los sistemas de inteligencia artificial (IA) han dado paso a una nueva era de abundancia de pruebas, pero ayudar a los humanos a comprender estas pruebas requiere una nueva atención a las diferencias en h

La noticia

arXiv:2611.06995v1 Announce Type: new Abstract: Artificial intelligence (AI) systems have ushered in a new age of proof abundance, but helping humans understand these proofs requires new attention to the differences in how AI agents and humans pursue a proof-making task. We present "simulated collaboration", a methodology that allows us to examine two key questions: (1) the relative attention AI agents pay to surface syntax or, conversely, deeper underlying semantics, and (2) how and where AI agent preferences for what to do next diverge from human preferences. On a set of nine major human collaborations in Lean with 4.7M definitions across more than 6K commits, we find that, over the course