Signal2026-07-16
Papers With Code

ShortOPD: Recovering Pruned LLMs with Short-to-Long On-Policy Distillation

Part of

Advances In Deep Learning Frameworks

Open primary sourceSee the whole picture