CO-Search: COVID-19 Information Retrieval with Semantic Search, Question Answering, and Abstractive Summarization

TL;DR

CO-Search is a semantic search engine tailored for COVID-19 literature, combining multiple retrieval and summarization techniques to assist health workers in quickly finding relevant scientific information during the pandemic.

Contribution

The paper introduces CO-Search, a novel retrieval system integrating semantic search, question answering, and summarization specifically designed for COVID-19 literature, with a unique training approach using citation graphs.

Findings

Achieved top performance on TREC-COVID datasets

Effective retrieval with high precision and recall metrics

Demonstrated utility in supporting health workers during COVID-19 crisis

Abstract

The COVID-19 global pandemic has resulted in international efforts to understand, track, and mitigate the disease, yielding a significant corpus of COVID-19 and SARS-CoV-2-related publications across scientific disciplines. As of May 2020, 128,000 coronavirus-related publications have been collected through the COVID-19 Open Research Dataset Challenge. Here we present CO-Search, a retriever-ranker semantic search engine designed to handle complex queries over the COVID-19 literature, potentially aiding overburdened health workers in finding scientific answers during a time of crisis. The retriever is built from a Siamese-BERT encoder that is linearly composed with a TF-IDF vectorizer, and reciprocal-rank fused with a BM25 vectorizer. The ranker is composed of a multi-hop question-answering module, that together with a multi-paragraph abstractive summarizer adjust retriever scores. To account for the domain-specific and relatively limited dataset, we generate a bipartite graph of document paragraphs and citations, creating 1.3 million (citation title, paragraph) tuples for training the encoder. We evaluate our system on the data of the TREC-COVID information retrieval challenge. CO-Search obtains top performance on the datasets of the first and second rounds, across several key metrics: normalized discounted cumulative gain, precision, mean average precision, and binary preference.