← back to paper

Review history

arxiv: 2608.09447 · 2 revisions

WDL-OPD: Weak-Driven On-Policy Distillation via Mixture-Constrained Co-Training

1. 2026-08-15 CONDITIONAL MODERATE v1.4.0-alltime-deepseek-medium novelty 6.0
   148137 ms 8836 in 19229 out 2026-08-15T14:24:23.588112+00:00
2. 2026-08-11 CONDITIONAL HIGH v1.4.0-daily-deepseek-medium novelty 6.0
   117782 ms 8851 in 12977 out 2026-08-11T17:18:05.631996+00:00