Textual logical reasoning, especially question-answering (QA) tasks with logical reasoning, requires awareness of particular logical structures. The passage-level logical relations represent entailment or contradiction between propositional units (e.g., a concluding sentence). However, such structures are unexplored as current QA systems focus on entity-based relations. In this work, we propose logic structural-constraint modeling to solve the logical reasoning QA and introduce discourse-aware graph networks (DAGNs). The networks first construct logic graphs leveraging in-line discourse connectives and generic logic theories, then learn logic representations by end-to-end evolving the logic relations with an edge-reasoning mechanism and updating the graph features. This pipeline is applied to a general encoder, whose fundamental features are joined with the high-level logic features for answer prediction. Experiments on three textual logical reasoning datasets demonstrate the reasonability of the logical structures built in DAGNs and the effectiveness of the learned logic features. Moreover, zero-shot transfer results show the features' generality to unseen logical texts.

翻译：文本逻辑推理，尤其是涉及逻辑推理的问答（QA）任务，需要感知特定的逻辑结构。段落层面的逻辑关系表示命题单元（如结论句）之间的蕴涵或矛盾关系。然而，当前问答系统主要关注基于实体的关系，这类结构尚未得到充分探索。在本工作中，我们提出逻辑结构约束建模来解决逻辑推理问答问题，并引入了话语感知图网络（DAGNs）。该网络首先利用在线话语连接词和通用逻辑理论构建逻辑图，然后通过端到端演化逻辑关系（借助边推理机制）并更新图特征来学习逻辑表示。这一流程应用于通用编码器，将其基础特征与高层逻辑特征相结合以进行答案预测。在三个文本逻辑推理数据集上的实验表明，DAGNs构建的逻辑结构具有合理性，且学习到的逻辑特征具有有效性。此外，零样本迁移结果进一步证明了这些特征对未见过的逻辑文本具有泛化性。