Multi-step Retriever-Reader Interaction for Scalable Open-domain Question Answering

TL;DR

This paper proposes a multi-step retriever-reader framework for open-domain question answering, enabling iterative interaction that improves retrieval and reading performance across large datasets.

Contribution

It introduces a novel multi-step interaction framework that is architecture-agnostic and scalable to large corpora, enhancing open-domain QA accuracy.

Findings

Iterative interaction improves paragraph retrieval quality.

Framework enhances performance of DrQA and BiDAF models.

Consistent gains across multiple large datasets.

Abstract

This paper introduces a new framework for open-domain question answering in which the retriever and the reader iteratively interact with each other. The framework is agnostic to the architecture of the machine reading model, only requiring access to the token-level hidden representations of the reader. The retriever uses fast nearest neighbor search to scale to corpora containing millions of paragraphs. A gated recurrent unit updates the query at each step conditioned on the state of the reader and the reformulated query is used to re-rank the paragraphs by the retriever. We conduct analysis and show that iterative interaction helps in retrieving informative paragraphs from the corpus. Finally, we show that our multi-step-reasoning framework brings consistent improvement when applied to two widely used reader architectures DrQA and BiDAF on various large open-domain datasets --- TriviaQA-unfiltered, QuasarT, SearchQA, and SQuAD-Open.