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

Open paper
Benchmark Model Rank Results
mathematical-reasoning-on-aime24R1-Distill-7B (Direct-OPD)#65Acc: 61.2
mathematical-reasoning-on-aime-2025R1-Distill-7B (Direct-OPD)#90Accuracy: 44.0

Rank counts only results with a code link.