Mira Murati (X)· miramurati·· 2025-10-28AI 评分36
同策略蒸馏结合RL与SFT优势
Combining the benefits of RL and SFT with on-policy distillation, a promising approach for training small models for domain performance a...
AI 导读
结合RL和SFT的优势，采用同策略蒸馏，这是一种有前景的方法，可用于训练小模型以提升领域表现和持续学习能力。
来源：Mira Murati (X) · x.com