期刊名称：Conference on European Chapter of the Association for Computational Linguistics (EACL)
出版年度：2021
卷号：2021
页码：1121-1133
DOI：10.18653/v1/2021.eacl-main.95
语种：English
出版社：ACL Anthology
摘要：In this work, we study how the finetuning stage in the pretrain-finetune framework changes the behavior of a pretrained neural language generator. We focus on the transformer encoder-decoder model for the open-domain dialogue response generation task. Our major finding is that after standard finetuning, the model forgets some of the important language generation skills acquired during large-scale pretraining. We demonstrate the forgetting phenomenon through a set of detailed behavior analysis from the perspectives of knowledge transfer, context sensitivity, and function space projection. As a preliminary attempt to alleviate the forgetting problem, we propose an intuitive finetuning strategy named “mix-review”. We find that mix-review effectively regularizes the finetuning process, and the forgetting problem is alleviated to some extent. Finally, we discuss interesting behavior of the resulting dialogue model and its implications.
