Technology & AIJun 29, 2026
DOPD: Dual On-policy Distillation
On-policy distillation (OPD) offers superior capacity transfer by supervising student-sampled trajectories with dense token-level signals.
On-policy distillation (OPD) offers superior capacity transfer by supervising student-sampled trajectories with dense token-level signals. To furnish high-quality supervision sources and thereby elevate the performance frontier of distillation, an intuitive direction is to…
Sign in to learn & save →
The frontier is open to all. Sign in to learn this from first principles and save it to your knowledge base.