Pith. sign in

← back to paper

Review history

arxiv: 2608.07935 · 3 revisions

Adaptive Supervised Anchoring for On-Policy Self-Distillation

  1. 2026-08-15 CONDITIONAL MODERATE v1.4.0-alltime-deepseek-medium novelty 6.0
    107214 ms 12226 in 13692 out 2026-08-15T14:25:49.571094+00:00
  2. 2026-08-12 CONDITIONAL MODERATE v1.4.0-alltime-deepseek-medium novelty 6.0
    106361 ms 12233 in 13457 out 2026-08-12T04:14:05.116855+00:00
  3. 2026-08-12 CONDITIONAL MODERATE v1.4.0-alltime-deepseek-medium novelty 5.0
    142694 ms 17857 in 14031 out 2026-08-12T00:39:10.754600+00:00