Ask Question First for Enhancing Lifelong Language Learning

Han Wang, Ruiliu Fu, Xuejun Zhang, Jun Zhou, Qingwei Zhao


Abstract
Lifelong language learning aims to stream learning NLP tasks while retaining knowledge of previous tasks. Previous works based on the language model and following data-free constraint approaches have explored formatting all data as “begin token (B) + context (C) + question (Q) + answer (A)” for different tasks. However, they still suffer from catastrophic forgetting and are exacerbated when the previous task’s pseudo data is insufficient for the following reasons: (1) The model has difficulty generating task-corresponding pseudo data, and (2) A is prone to error when A and C are separated by Q because the information of the C is diminished before generating A. Therefore, we propose the Ask Question First and Replay Question (AQF-RQ), including a novel data format “BQCA” and a new training task to train pseudo questions of previous tasks. Experimental results demonstrate that AQF-RQ makes it easier for the model to generate more pseudo data that match corresponding tasks, and is more robust to both sufficient and insufficient pseudo-data when the task boundary is both clear and unclear. AQF-RQ can achieve only 0.36% lower performance than multi-task learning.
Anthology ID:
2022.coling-1.408
Volume:
Proceedings of the 29th International Conference on Computational Linguistics
Month:
October
Year:
2022
Address:
Gyeongju, Republic of Korea
Venue:
COLING
SIG:
Publisher:
International Committee on Computational Linguistics
Note:
Pages:
4610–4621
Language:
URL:
https://aclanthology.org/2022.coling-1.408
DOI:
Bibkey:
Cite (ACL):
Han Wang, Ruiliu Fu, Xuejun Zhang, Jun Zhou, and Qingwei Zhao. 2022. Ask Question First for Enhancing Lifelong Language Learning. In Proceedings of the 29th International Conference on Computational Linguistics, pages 4610–4621, Gyeongju, Republic of Korea. International Committee on Computational Linguistics.
Cite (Informal):
Ask Question First for Enhancing Lifelong Language Learning (Wang et al., COLING 2022)
Copy Citation:
PDF:
https://aclanthology.org/2022.coling-1.408.pdf
Data
decaNLP