SPARTA: Efficient Open-Domain Question Answering via Sparse Transformer Matching Retrieval

Abstract

We introduce SPARTA, a novel neural retrieval method that shows great promise in performance, generalization, and interpretability for open-domain question answering. Unlike many neural ranking methods that use dense vector nearest neighbor search, SPARTA learns a sparse representation that can be efficiently implemented as an Inverted Index. The resulting representation enables scalable neural retrieval that does not require expensive approximate vector search and leads to better performance than its dense counterpart. We validated our approaches on 4 open-domain question answering (OpenQA) tasks and 11 retrieval question answering (ReQA) tasks. SPARTA achieves new state-of-the-art results across a variety of open-domain question answering tasks in both English and Chinese datasets, including open SQuAD, Natuarl Question, CMRC and etc. Analysis also confirms that the proposed method creates human interpretable representation and allows flexible control over the trade-off between performance and efficiency.

11 pages

References in corpus (8)

• A Deep Relevance Matching Model for Ad-hoc Retrieval
• End-to-End Neural Ad-hoc Ranking with Kernel Pooling
• Learning to Retrieve Reasoning Paths over Wikipedia Graph for Question Answering
• Pre-training Tasks for Embedding-based Large-scale Retrieval
• DROP: A Reading Comprehension Benchmark Requiring Discrete Reasoning Over Paragraphs
• Multi-passage BERT: A Globally Normalized BERT Model for Open-domain Question Answering
• MRQA 2019 Shared Task: Evaluating Generalization in Reading Comprehension
• Retrieve, Read, Rerank: Towards End-to-End Multi-Document Reading Comprehension