We present an accurate and interpretable method for answer extraction in machine reading comprehension that is reminiscent of case-based reasoning (CBR) from classical AI. Our method (CBR-MRC) builds upon the hypothesis that contextualized answers to similar questions share semantic similarities with each other. Given a test question, CBR-MRC first retrieves a set of similar cases from a nonparametric memory and then predicts an answer by selecting the span in the test context that is most similar to the contextualized representations of answers in the retrieved cases. The semi-parametric nature of our approach allows it to attribute a prediction to the specific set of evidence cases, making it a desirable choice for building reliable and debuggable QA systems. We show that CBR-MRC provides high accuracy comparable with large reader models and outperforms baselines by 11.5 and 8.4 EM on NaturalQuestions and NewsQA, respectively. Further, we demonstrate the ability of CBR-MRC in identifying not just the correct answer tokens but also the span with the most relevant supporting evidence. Lastly, we observe that contexts for certain question types show higher lexical diversity than others and find that CBR-MRC is robust to these variations while performance using fully-parametric methods drops.

翻译：我们提出了一种精准且可解释的机器阅读理解答案抽取方法，该方法继承了经典人工智能中基于案例推理（CBR）的思想。我们的方法（CBR-MRC）基于以下假设：相似问题的上下文化答案之间存在语义相似性。针对测试问题，CBR-MRC首先从非参数化记忆库中检索一组相似案例，然后通过选择测试上下文中最相似于检索案例中答案上下文化表示的文本片段来预测答案。该方法半参数化的特性使其能够将预测归因于特定证据案例集，成为构建可靠且可调试问答系统的理想选择。实验表明，CBR-MRC在保持与大规模阅读器模型相当的高准确率的同时，在NaturalQuestions和NewsQA数据集上分别以11.5和8.4的精确匹配（EM）分数超越基线方法。此外，我们证明了CBR-MRC不仅能够识别正确答案标记，还能定位最具相关支持证据的文本片段。最后，我们发现某些类型问题的上下文在词汇多样性上表现显著，并观察到CBR-MRC对此类变化具有鲁棒性，而全参数化方法的性能则会下降。