arxiv
PublishedApril 11, 2026 at 4:00 AM
—neutral
Demystifying OPD: Length Inflation and Stabilization Strategies for Large Language Models
Publisher summary· verbatim
arXiv:2604.08527v1 Announce Type: new Abstract: On-policy distillation (OPD) trains student models under their own induced distribution while leveraging supervision from stronger teachers. We identify a failure mode of OPD: as training progresses, on-policy rollouts can undergo abrupt length inflati
Stay posted· Newsletter
A 5-min weekly brief — top movers, price watch, story of the week.
Discussion
No replies yet. Be first.
The Bubble Brief
WEEKLYRead AI insights every Tuesday — top movers, new releases, story of the week.
Originally published on arxiv ↗