NYU Debate Training Update: Methods, Baselines, Preliminary Results 2024-07-06T18:28:54.053Z

Comment by samarnesen on GPT-3.5 judges can supervise GPT-4o debaters in capability asymmetric debates · 2024-08-28T03:25:04.183Z · LW · GW

This seems like really interesting work!  Would you be able to share any example transcripts from some of these debates? Since RLHF'ed models often shy away from combativeness, I'm curious as to the form of GPT-4's rebuttals (especially for questions where the judge gets it right after reading the debate but wrong otherwise)