blueblaze0
Re:Prover-Verifier Games improve legibility of language model outputs
on Wednesday July 17, 2024 @08:50PM (#40991608)
This is impressive, but what prevents the blue agent from generating an incorrect proof of a "true example"? What prevents the red agent from generating a correct disproof of a "false example"? I'm curious how they managed to generate a truly unlimited source of correctly labeled examples.