1 code implementation • EMNLP 2021 • Erguang Yang, Mingtong Liu, Deyi Xiong, Yujie Zhang, Yao Meng, Changjian Hu, Jinan Xu, Yufeng Chen
Particularly, we design a two-stage learning method to effectively train the model using non-parallel data.
no code implementations • CCL 2021 • Bo Jin, Mingtong Liu, Yujie Zhang, Jinan Xu, Yufeng Chen
“如何挖掘语言资源中丰富的复述模板, 是复述研究中的一项重要任务。已有方法在人工给定种子实体对的基础上, 利用实体关系, 通过自举迭代方式, 从开放域获取复述模板, 规避对平行语料或可比语料的依赖, 但是该方法需人工给定实体对, 实体关系受限;在迭代过程中语义会发生偏移, 影响获取质量。针对这些问题, 我们考虑知识库中包含描述特定语义关系的实体对(即关系三元组), 提出融合外部知识的开放域复述模板自动获取方法。首先, 将关系三元组与开放域文本对齐, 获取关系对应文本, 并将文本中语义丰富部分泛化成变量槽, 获取关系模板;接着设计模板表示方法, 本文利用预训练语言模型, 在模板表示中融合变量槽语义;最后, 根据获得的模板表示, 设计自动聚类与筛选方法, 获取高精度的复述模板。在融合自动评测与人工评测的评价方法下, 实验结果表明, 本文提出的方法实现了在开放域数据上复述模板的自动泛化与获取, 能够获得质量高、语义一致的复述模板。”
no code implementations • CCL 2020 • Xingchen Li, Mingtong Liu, Yujie Zhang, Jinan Xu, Yufeng Chen
The experimental results on the Penn Chinese treebank (CTB5) show that our proposed joint model improved by 0. 38% on dependency parsing than the model of Yan et al. (2019).
no code implementations • COLING 2020 • Mingtong Liu, Erguang Yang, Deyi Xiong, Yujie Zhang, Yao Meng, Changjian Hu, Jinan Xu, Yufeng Chen
We propose a learning-exploring method to generate sentences as learning objectives from the learned data distribution, and employ reinforcement learning to combine these new learning objectives for model training.
no code implementations • IJCNLP 2019 • Mingtong Liu, Yu-Jie Zhang, Jinan Xu, Yufeng Chen
Unlike existing models, each attention layer of OSOA-DFN is oriented to the original semantic representation of another sentence, which captures the relevant information from a fixed matching target.