Towards Robustness and Diversity: Continual Learning in Dialog Generation with Text-Mixup and Batch Nuclear-Norm Maximization
Fuente:
arXiv
Salvato in:
| Autori principali: | , , , , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2024
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866914716193914880 |
|---|---|
| author | Wang, Zihan Xiao, Jiayu Li, Mengxiang He, Zhongjiang Li, Yongxiang Wang, Chao Song, Shuangyong |
| author_facet | Wang, Zihan Xiao, Jiayu Li, Mengxiang He, Zhongjiang Li, Yongxiang Wang, Chao Song, Shuangyong |
| contents | In our dynamic world where data arrives in a continuous stream, continual learning enables us to incrementally add new tasks/domains without the need to retrain from scratch. A major challenge in continual learning of language model is catastrophic forgetting, the tendency of models to forget knowledge from previously trained tasks/domains when training on new ones. This paper studies dialog generation under the continual learning setting. We propose a novel method that 1) uses \textit{Text-Mixup} as data augmentation to avoid model overfitting on replay memory and 2) leverages Batch-Nuclear Norm Maximization (BNNM) to alleviate the problem of mode collapse. Experiments on a $37$-domain task-oriented dialog dataset and DailyDialog (a $10$-domain chitchat dataset) demonstrate that our proposed approach outperforms the state-of-the-art in continual learning. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2403_10894 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Towards Robustness and Diversity: Continual Learning in Dialog Generation with Text-Mixup and Batch Nuclear-Norm Maximization Wang, Zihan Xiao, Jiayu Li, Mengxiang He, Zhongjiang Li, Yongxiang Wang, Chao Song, Shuangyong Computation and Language In our dynamic world where data arrives in a continuous stream, continual learning enables us to incrementally add new tasks/domains without the need to retrain from scratch. A major challenge in continual learning of language model is catastrophic forgetting, the tendency of models to forget knowledge from previously trained tasks/domains when training on new ones. This paper studies dialog generation under the continual learning setting. We propose a novel method that 1) uses \textit{Text-Mixup} as data augmentation to avoid model overfitting on replay memory and 2) leverages Batch-Nuclear Norm Maximization (BNNM) to alleviate the problem of mode collapse. Experiments on a $37$-domain task-oriented dialog dataset and DailyDialog (a $10$-domain chitchat dataset) demonstrate that our proposed approach outperforms the state-of-the-art in continual learning. |
| title | Towards Robustness and Diversity: Continual Learning in Dialog Generation with Text-Mixup and Batch Nuclear-Norm Maximization |
| topic | Computation and Language |
| url | https://arxiv.org/abs/2403.10894 |