DAGN: Discourse-Aware Graph Network for Logical Reasoning
Yinya Huang1 Meng Fang2 Yu Cao3 Liwei Wang4 Xiaodan Liang1 1Shenzhen Campus of Sun Yat-sen University 2Tencent Robotics X 3School of Computer Science, The University of Sydney 4The Chinese University of Hong Kong This work was done during Yinya Huang’s internship in Tencent with L. Wang and M. Fang. Corresponding Author: Xiaodan Liang.
Abstract
Recent QA with logical reasoning questions requires passage-level relations among the sentences. However, current approaches still focus on sentence-level relations interacting among tokens. In this work, we explore aggregating passage-level clues for solving logical reasoning QA by using discourse-based information. We propose a discourse-aware graph network (DAGN) that reasons relying on the discourse structure of the texts. The model encodes discourse information as a graph with elementary discourse units (EDUs) and discourse relations, and learns the discourse-aware features via a graph network for downstream QA tasks. Experiments are conducted on two logical reasoning QA datasets, ReClor and LogiQA, and our proposed DAGN achieves competitive results. The source code is available at https://github.com/Eleanor-H/DAGN.
中文速览
逻辑推理类问答任务要求模型理解段落中句子之间的深层逻辑关系，但现有方法大多停留在词级别的交互，无法有效捕捉段落级别的逻辑结构。为此，研究者提出了一种名为"话语感知图网络"（Discourse-Aware Graph Network，DAGN）的方法：先利用PDTB 2.0中的话语连接词（如"because""while"等）和标点符号，将文本切分为基本话语单元（Elementary Discourse Units，EDUs），再以EDU为节点、话语关系为边构建逻辑图，最后通过图神经网络学习话语级特征，并将其与预训练语言模型（RoBERTa-Large）的上下文特征融合，用于多选问答的答案预测。在ReClor和LogiQA两个逻辑推理问答数据集上，DAGN均超越了基线模型，尤其在ReClor的难题子集上比RoBERTa-Large提升了4.46%，在LogiQA测试集上提升了4.01%。这项工作表明，将话语结构显式地引入逻辑推理模型是一条切实可行的路径，为机器理解文本深层逻辑提供了新思路。
原文 arXiv:2103.14349;中英对照 + 大白话阅读 https://aha.fim.ai/paper/2103.14349v2