When Retriever-Reader Meets Scenario-Based Multiple-Choice Questions
This addresses the problem of improving retrieval and reading for scenario-based QA, which is important for applications requiring contextual understanding, but it is incremental as it builds on existing retriever-reader frameworks.
The paper tackles the challenge of scenario-based question answering (SQA), where retrieval is difficult due to noisy scenario descriptions and lack of supervision, by proposing a joint retriever-reader model called JEEVES that uses implicit supervision via word weighting. The result is that JEEVES significantly outperforms strong baselines on multiple-choice questions across three SQA datasets.
Scenario-based question answering (SQA) requires retrieving and reading paragraphs from a large corpus to answer a question which is contextualized by a long scenario description. Since a scenario contains both keyphrases for retrieval and much noise, retrieval for SQA is extremely difficult. Moreover, it can hardly be supervised due to the lack of relevance labels of paragraphs for SQA. To meet the challenge, in this paper we propose a joint retriever-reader model called JEEVES where the retriever is implicitly supervised only using QA labels via a novel word weighting mechanism. JEEVES significantly outperforms a variety of strong baselines on multiple-choice questions in three SQA datasets.