Pith. sign in

← back to paper

Review history

arxiv: 2607.05394 · 2 revisions

Weak-to-Strong Generalization via Direct On-Policy Distillation

  1. 2026-07-11 CONDITIONAL HIGH v1.1.0-grok45 novelty 6.0
    32685 ms 27139 in 3809 out 2026-07-11T07:01:00.230633+00:00
  2. 2026-07-07 CONDITIONAL MODERATE v1.1.0-glm novelty 7.0
    191408 ms 24949 in 2529 out 2026-07-07T12:22:06.016759+00:00