2 papers
cs.LG2026
Test-Time Self-Play: Scaling without a Verifier
Akira Yoshiyama, Akmal Ashirmatov, Jonas Hübotter +2
Test-time scaling methods spend additional compute at deployment to improve a large language model's performance on a hard target problem. Most methods rely on a verifier that scor…
cs.LG2026
On Repulsive and Attractive Teachers: Separating Correctness from Behavior in Self-Distillation
Anton Baumann, Akmal Ashirmatov, Leo Schmidt-Traub +4
On-policy self-distillation provides dense, token-level supervision by conditioning a model on privileged information and distilling the resulting teacher distribution back into th…