Two papers are accepted by ACL 2020!

One is about multidomain NMT using word level mixing model.

The other one is SMART model, leaderboad wining model on GLUE.

Reference⌗

  • SMART: Robust and Efficient Fine-Tuning for Pre-trained Natural Language Models through Principled Regularized Optimization 2020
    Haoming Jiang, Pengcheng He, Weizhu Chen, Xiaodong Liu, Jianfeng Gao and Tuo Zhao [arXiv] [Code]
    Annual Conference of the Association for Computational Linguistics (ACL), 2020

  • Multi-Domain Neural Machine Translation with Word-Level Adaptive Layer-wise Domain Mixing 2020
    Haoming Jiang, Chen Liang, Chong Wang and Tuo Zhao [arXiv]
    Annual Conference of the Association for Computational Linguistics (ACL), 2020