论文标题
歌剧：文本上的行动离散推理
OPERA:Operation-Pivoted Discrete Reasoning over Text
论文作者
论文摘要
机器阅读理解（MRC）需要涉及符号操作的离散推理，例如加法，排序和计数，这是一项具有挑战性的任务。根据这种性质，基于语义解析的方法可以预测可解释但复杂的逻辑形式。但是，逻辑形式的生成是不平凡的，即使以逻辑形式有些扰动也会导致错误的答案。为了减轻此问题，提出了基于多预测的方法来直接预测不同类型的答案并取得改进。但是，他们忽略了符号操作的利用，而遇到缺乏推理能力和解释性。为了继承这两种方法的优势，我们提出了Opera，这是一个欺骗性的离散推理框架，其中轻巧的符号操作（与逻辑形式相比）作为神经模块可用于促进推理能力和解释性。具体而言，首先选择操作，然后轻轻执行以模拟答案推理过程。关于Drop和Racenum数据集的广泛实验表明了歌剧的推理能力。此外，进一步的分析验证其可解释性。
Machine reading comprehension (MRC) that requires discrete reasoning involving symbolic operations, e.g., addition, sorting, and counting, is a challenging task. According to this nature, semantic parsing-based methods predict interpretable but complex logical forms. However, logical form generation is nontrivial and even a little perturbation in a logical form will lead to wrong answers. To alleviate this issue, multi-predictor -based methods are proposed to directly predict different types of answers and achieve improvements. However, they ignore the utilization of symbolic operations and encounter a lack of reasoning ability and interpretability. To inherit the advantages of these two types of methods, we propose OPERA, an operation-pivoted discrete reasoning framework, where lightweight symbolic operations (compared with logical forms) as neural modules are utilized to facilitate the reasoning ability and interpretability. Specifically, operations are first selected and then softly executed to simulate the answer reasoning procedure. Extensive experiments on both DROP and RACENum datasets show the reasoning ability of OPERA. Moreover, further analysis verifies its interpretability.
