概念
distillation training
创建2026-04-15
更新2026-04-15
阅读量级1 分钟
概念导读

Strong-to-weak distillation is a paradigm where a larger, more capable "strong" teacher model distills its knowledge into a smaller, less capable "weak" student model. This is the most common distillation direction and is used in models like qwen3 and frameworks like generalized-on-policy-distillation. The goal is to preserve as much of the strong model's ca...

  1. Related
  2. References

Strong-to-Weak Distillation

Strong-to-weak distillation is a paradigm where a larger, more capable "strong" teacher model distills its knowledge into a smaller, less capable "weak" student model. This is the most common distillation direction and is used in models like qwen3 and frameworks like generalized-on-policy-distillation. The goal is to preserve as much of the strong model's capability as possible in the weaker student, creating efficient models for deployment.

References