Revisiting Knowledge Distillation for Autoregressive Language Models
Qihuang Zhong, Liang Ding, Li Shen, Juhua Liu, Bo Du, Dacheng Tao · 2024
Knowledge distillation (KD) is a common approach to compress a teacher model to reduce its inference cost and memory footprint, by training a smaller student model.However, in the context of autoregressive language models (LMs), we empirically find that larger teachers might dramatically result in a poorer student.In response to this problem, we conduct a series of analyses and reveal that different tokens have different teaching modes, neglecting which will lead to performance degradation.Motivated by this, we propose a simple yet effective adaptive teaching approach (ATKD) to improve the KD.The core of ATKD is to reduce rote learning and make teaching more diverse and flexible.Extensive experiments on 8 LM tasks show that, with the help of ATKD, various baseline KD methods can achieve consistent and significant performance gains (up to +3.04% average score) across all model types and sizes.More encouragingly, ATKD can improve the student model generalization effectively.