no code implementations • EMNLP 2020 • Chen Jia, Yuefeng Shi, Qinrong Yang, Yue Zhang
We then integrate the entity information into BERT using Char-Entity-Transformer, which augments the self-attention using a combination of character and entity representations.
no code implementations • 1 Mar 2024 • Jinman Zhao, Yitian Ding, Chen Jia, Yining Wang, Zifan Qian
We investigate the outputs of the GPT series of LLMs in various languages using our three measurement methods.
no code implementations • 22 Feb 2024 • Chen Jia
During meta-training, a bilevel optimization algorithm is utilized to learn a reward model capable of guiding policy learning to align with human preferences across various distributions.
no code implementations • 29 Sep 2021 • Chen Jia
This work attempts to tackle the problem of domain generalization (DG) via learning to reduce domain shift with an episodic training procedure.
no code implementations • 28 Jun 2021 • Qingjian Lin, Lin Yang, Xuyang Wang, Luyuan Xie, Chen Jia, Junjie Wang
This paper proposes the weighted SI-SNR loss, together with the joint learning of target speech separation and personal VAD.
no code implementations • 9 Feb 2021 • Youming Li, Da-Quan Jiang, Chen Jia
Here we develop a novel method of computing the exact joint distributions of a wide class of first-order stochastic reaction systems in steady-state conditions.
no code implementations • ACL 2020 • Chen Jia, Yue Zhang
Cross-domain NER is a challenging yet practical problem.
no code implementations • 8 Jan 2020 • Chen Jia, Ramon Grima
Furthermore we show that our model predictions for the protein number distribution are significantly different from those of Kumar et al. when the protein mean is small, gene switching is fast, and protein binding is faster than unbinding.
no code implementations • 21 Sep 2019 • Chen Jia, Hong Qian, Michael Q. Zhang
In our model, oscillations tend to occur when the protein is relatively stable and when gene switching is relatively slow.
1 code implementation • ACL 2019 • Chen Jia, Xiaobo Liang, Yue Zhang
Due to limitation of labeled resources, cross-domain named entity recognition (NER) has been a challenging task.