1 code implementation • COLING 2022 • Han Wang, Ruiliu Fu, Xuejun Zhang, Jun Zhou, Qingwei Zhao
Lifelong language learning aims to stream learning NLP tasks while retaining knowledge of previous tasks.
1 code implementation • 22 May 2022 • Han Wang, Ruiliu Fu, Xuejun Zhang, Jun Zhou
In order to alleviate catastrophic forgetting, we propose the residual variational autoencoder (RVAE) to enhance LAMOL, a recent LLL model, by mapping different tasks into a limited unified semantic space.
1 code implementation • Findings (EMNLP) 2021 • Ruiliu Fu, Han Wang, Xuejun Zhang, Jun Zhou, Yonghong Yan
The Relation Extractor decomposes the complex question, and then the Reader answers the sub-questions in turn, and finally the Comparator performs numerical comparison and summarizes all to get the final answer, where the entire process itself constitutes a complete reasoning evidence path.
no code implementations • 17 Oct 2021 • Han Wang, Ruiliu Fu, Chengzhang Li, Xuejun Zhang, Jun Zhou, Yonghong Yan
Incremental language learning with pseudo-data can alleviate catastrophic forgetting in neural networks.