Analyzing the Forgetting Problem in the Pretrain-Finetuning of Dialogue Response Models

16 Oct 2019  ·  Tianxing He, Jun Liu, Kyunghyun Cho, Myle Ott, Bing Liu, James Glass, Fuchun Peng ·

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. read more

PDF Abstract

Datasets


Results from the Paper


  Submit results from this paper to get state-of-the-art GitHub badges and help the community compare results to other papers.

Methods