When Retriever-Reader Meets Scenario-Based Multiple-Choice Questions

TL;DR

This paper introduces JEEVES, a joint retriever-reader model for scenario-based multiple-choice question answering that effectively retrieves relevant information without explicit supervision, significantly improving performance.

Contribution

The paper proposes a novel joint retriever-reader model with implicit supervision through QA labels, addressing retrieval challenges in SQA without needing relevance labels.

Findings

JEEVES outperforms strong baselines on multiple SQA datasets.

The model effectively handles noisy and keyphrase-rich scenarios.

Implicit supervision via QA labels enhances retrieval accuracy.

Abstract

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.