Weak-to-Strong Generalization via Direct On-Policy Distillation
AI 摘要
通过直接在线策略蒸馏实现弱到强泛化
Weak-to-Strong Generalization via Direct On-Policy Distillation
通过直接在线策略蒸馏实现弱到强泛化
Weak-to-Strong Generalization via Direct On-Policy Distillation
通过直接在线策略蒸馏实现弱到强泛化
Weak-to-Strong Generalization via Direct On-Policy Distillation