面向机器阅读理解的事实驱动逻辑推理
Fact-driven Logical Reasoning for Machine Reading Comprehension
AI总结：
本文提出一种事实驱动的层次化知识表示方法，通过提取句子主干构成事实单元并构建超图，同时建模句子级和实体级交互，在逻辑推理与对话建模任务上显著提升多种基线模型性能。
摘要：
近年来，训练具备推理能力的机器受到越来越多的关注，而推理能力在很大程度上依赖于准确且清晰呈现的线索形式。在现有研究中，线索通常被建模为实体感知知识。然而，这些实体感知线索主要关注常识，因此对于需要临时事实或事件知识的任务而言不够充分，尤其是在阅读理解的逻辑推理中。为了解决这一挑战，我们致力于以层次化的方式同时覆盖常识和临时知识线索。具体来说，我们通过提取句子的主干成分（例如构成“事实”的主语-动词-宾语结构），提出了一种通用的知识单元形式化表示。随后，我们在事实单元之上构建了一个超图，从而能够同时利用句子级交互（事实组之间的关系）和实体级交互（事实内部的概念或动作）。在逻辑推理基准和对话建模数据集上的实验结果表明，我们的方法显著提升了基线模型的表现，并且在不同的骨干模型上具有通用性。代码可在 \url{https://github.com/ozyyshr/FocalReasoner} 获取。
英文摘要：
Recent years have witnessed an increasing interest in training machines with reasoning ability, which deeply relies on accurately and clearly presented clue forms. The clues are usually modeled as entity-aware knowledge in existing studies. However, those entity-aware clues are primarily focused on commonsense, making them insufficient for tasks that require knowledge of temporary facts or events, particularly in logical reasoning for reading comprehension. To address this challenge, we are motivated to cover both commonsense and temporary knowledge clues hierarchically. Specifically, we propose a general formalism of knowledge units by extracting backbone constituents of the sentence, such as the subject-verb-object formed ``facts''. We then construct a supergraph on top of the fact units, allowing for the benefit of sentence-level (relations among fact groups) and entity-level interactions (concepts or actions inside a fact). Experimental results on logical reasoning benchmarks and dialogue modeling datasets show that our approach improves the baselines substantially, and it is general across backbone models. Code is available at \url{https://github.com/ozyyshr/FocalReasoner}.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
