OISD: On-Policy Internal Self-Distillation of Language Models 事件

OISD: On-Policy Internal Self-Distillation of Language Models arXiv:2605.29089v1 Announce Type: cross Abstract: Recent reinforcement learning (RL) post-training approaches primarily optimize the final output policy using sparse outcome-level rewards, while largely overlooking predictive signals encoded in intermediate representations. In this paper, we introduce a new paradigm called on-policy internal self-distillation and propose the OISD framework, which improves reasoning by transferring on