← back to paper

Review history

arxiv: 2605.09725 · 2 revisions

On-Policy Distillation with Best-of-N Teacher Rollout Selection

1. 2026-05-14 UNVERDICTED LOW v0.9.0 novelty 5.0
   28803 ms 5616 in 1135 out 2026-05-14T21:09:13.233274+00:00
2. 2026-05-12 UNVERDICTED LOW v0.9.0 novelty 5.0
   66626 ms 5616 in 1125 out 2026-05-12T02:28:10.255192+00:00