Evidence Aggregation for Answer Re-Ranking in Open-Domain Question Answering

Abstract

A popular recent approach to answering open-domain questions is to first search for question-related passages and then apply reading comprehension models to extract answers. Existing methods usually extract answers from single passages independently. But some questions require a combination of evidence from across different sources to answer correctly. In this paper, we propose two models which make use of multiple passages to generate their answers. Both use an answer-reranking approach which reorders the answer candidates generated by an existing state-of-the-art QA model. We propose two methods, namely, strength-based re-ranking and coverage-based re-ranking, to make use of the aggregated evidence from different passages to better determine the answer. Our models have achieved state-of-the-art results on three public open-domain QA datasets: Quasar-T, SearchQA and the open-domain version of TriviaQA, with about 8 percentage points of improvement over the former two datasets.

13 pages

References in corpus (5)

• Large-scale Simple Question Answering with Memory Networks
• SearchQA: A New Q&A Dataset Augmented with Context from a Search Engine
• Quasar: Datasets for Question Answering by Search and Reading
• R${}^3$: Reinforced Reader-Ranker for Open-Domain Question Answering
• S-Net: From Answer Extraction to Answer Generation for Machine Reading Comprehension

Cited by in corpus (33)

• Retrieving and Reading: A Comprehensive Survey on Open-domain Question Answering
• K-Adapter: Infusing Knowledge into Pre-Trained Models with Adapters
• Automated Fact Checking: Task formulations, methods and future directions
• Exploring Graph-structured Passage Representation for Multi-hop Reading Comprehension with Graph Neural Networks
• Representation Learning for Natural Language Processing
• Entity-Relation Extraction as Multi-Turn Question Answering
• Retrieve-and-Read: Multi-task Learning of Information Retrieval and Reading Comprehension
• Putting Question-Answering Systems into Practice: Transfer Learning for Efficient Domain Customization
• Neural Machine Reading Comprehension: Methods and Trends
• A Co-Matching Model for Multi-choice Reading Comprehension
• Simple and Effective Curriculum Pointer-Generator Networks for Reading Comprehension over Long Narratives
• Improving Natural Language Inference Using External Knowledge in the Science Questions Domain
• Ranking Paragraphs for Improving Answer Recall in Open-Domain Question Answering
• Open Domain Question Answering Using Early Fusion of Knowledge Bases and Text
• Multi-Passage Machine Reading Comprehension with Cross-Passage Answer Verification
• Joint Training of Candidate Extraction and Answer Selection for Reading Comprehension
• Multi-Stage Conversational Passage Retrieval: An Approach to Fusing Term Importance Estimation and Neural Query Rewriting
• Knowledge-Aided Open-Domain Question Answering
• Learning to Coordinate Multiple Reinforcement Learning Agents for Diverse Query Reformulation
• RECONSIDER: Re-Ranking using Span-Focused Cross-Attention for Open Domain Question Answering
• A Multi-Type Multi-Span Network for Reading Comprehension that Requires Discrete Reasoning
• Finding Generalizable Evidence by Learning to Convince Q&A Models
• Learning to Search in Long Documents Using Document Structure
• Progressively Pretrained Dense Corpus Index for Open-Domain Question Answering
• Answering Open-Domain Questions of Varying Reasoning Steps from Text
• Multi-grained Evidence Inference for Multi-choice Reading Comprehension
• Complementary Evidence Identification in Open-Domain Question Answering
• Answering Any-hop Open-domain Questions with Iterative Document Reranking
• Interactive Fiction Game Playing as Multi-Paragraph Reading Comprehension with Reinforcement Learning
• Syntax-Enhanced Pre-trained Model
• Learning Representations and Agents for Information Retrieval
• A Study of the Tasks and Models in Machine Reading Comprehension
• Dual Reader-Parser on Hybrid Textual and Tabular Evidence for Open Domain Question Answering