TransformerG:基于层级图结构与文本注意力机制的法律文本多跳阅读理解

TransformerG:Multi-hop Reading Comprehension of Legal Texts Based on Hierarchical Graph Structure and Attention Mechanism

该文针对Cail2020法律多跳机器阅读理解数据集进行研究,提出了TransformerG,一个基于不同层级的实体图结构与文本信息的注意力机制融合的多跳阅读理解模型。该模型有效地结合了段落中问题节点、问题的实体节点、句子节点、句中的实体节点的特征与文本信息的特征,从而预测答案片段。此外,该文提出了一种句子级滑动窗口的方法,有效解决在预训练模型中文本过长导致的截断问题。利用TransformerG模型参加中国中文信息学会计算语言学专委会(CIPS-CL)和最高人民法院信息中心举办的“中国法研杯”司法人工智能挑战赛机器阅读理解赛道,取得了第2名的成绩。

Focused on the Cail2020 multi-hop machine reading comprehension data set,this paper presents TransformerG,a multi-hop reading comprehension model based on the integration of paragraph graph structure and attention mechanism.This model combines the feature of question node,question entity node,sentence node and sentence entity node in the text to predict the answer span.In addition,a sentence level sliding window method is designed to substitute the truncation of long text in the pre-training model.The proposed TransformerG model ranks Top 2 in the machine reading comprehension setting of Cail2020 Competition.