Retrieving and Reading: A Comprehensive Survey on Open-domain Question Answering

Abstract

Open-domain Question Answering (OpenQA) is an important task in Natural Language Processing (NLP), which aims to answer a question in the form of natural language based on large-scale unstructured documents. Recently, there has been a surge in the amount of research literature on OpenQA, particularly on techniques that integrate with neural Machine Reading Comprehension (MRC). While these research works have advanced performance to new heights on benchmark datasets, they have been rarely covered in existing surveys on QA systems. In this work, we review the latest research trends in OpenQA, with particular attention to systems that incorporate neural MRC techniques. Specifically, we begin with revisiting the origin and development of OpenQA systems. We then introduce modern OpenQA architecture named "Retriever-Reader" and analyze the various systems that follow this architecture as well as the specific techniques adopted in each of the components. We then discuss key challenges to developing OpenQA systems and offer an analysis of benchmarks that are commonly used. We hope our work would enable researchers to be informed of the recent advancement and also the open challenges in OpenQA research, so as to stimulate further progress in this field.

References in corpus (27)

• Exploring the Limits of Transfer Learning with a Unified Text-to-Text Transformer
• Retrieval-Augmented Generation for Knowledge-Intensive NLP Tasks
• Language Models are Few-Shot Learners
• Teaching Machines to Read and Comprehend
• XLNet: Generalized Autoregressive Pretraining for Language Understanding
• Bidirectional Attention Flow for Machine Comprehension
• REALM: Retrieval-Augmented Language Model Pre-Training
• The Goldilocks Principle: Reading Children's Books with Explicit Memory Representations
• Think you have Solved Question Answering? Try ARC, the AI2 Reasoning Challenge
• End-to-End Open-Domain Question Answering with BERTserini
• SearchQA: A New Q&A Dataset Augmented with Context from a Search Engine
• Advances and Challenges in Conversational Recommender Systems: A Survey
• ReCoRD: Bridging the Gap between Human and Machine Commonsense Reading Comprehension
• Evidence Aggregation for Answer Re-Ranking in Open-Domain Question Answering
• Language Models as Knowledge Bases?
• Knowledge Guided Text Retrieval and Reading for Open Domain Question Answering
• Recent Advances in Neural Question Generation
• Explicit State Tracking with Semi-Supervision for Neural Dialogue Generation
• Differentiable Reasoning over a Virtual Knowledge Base
• Neural Question Generation from Text: A Preliminary Study
• Multi-step Retriever-Reader Interaction for Scalable Open-domain Question Answering
• Generation-Augmented Retrieval for Open-domain Question Answering
• Read + Verify: Machine Reading Comprehension with Unanswerable Questions
• Answering Complex Open-Domain Questions with Multi-Hop Dense Retrieval
• DC-BERT: Decoupling Question and Document for Efficient Contextual Encoding
• SPARTA: Efficient Open-Domain Question Answering via Sparse Transformer Matching Retrieval
• Answering Any-hop Open-domain Questions with Iterative Document Reranking

Cited by in corpus (21)

• Advances and Challenges in Conversational Recommender Systems: A Survey
• QA Dataset Explosion: A Taxonomy of NLP Resources for Question Answering and Reading Comprehension
• QAnswer: Towards Question Answering Search over Websites
• Exploring the State of the Art in Legal QA Systems
• MultiDoc2Dial: Modeling Dialogues Grounded in Multiple Documents
• Multimodal Neural Databases
• Towards Standard Criteria for human evaluation of Chatbots: A Survey
• TAT-QA: A Question Answering Benchmark on a Hybrid of Tabular and Textual Content in Finance
• More Than Reading Comprehension: A Survey on Datasets and Metrics of Textual Question Answering
• Building Efficient and Effective OpenQA Systems for Low-Resource Languages
• Large-Scale Knowledge Synthesis and Complex Information Retrieval from Biomedical Documents
• Evaluating the Knowledge Dependency of Questions
• Generalized Relation Learning with Semantic Correlation Awareness for Link Prediction
• Dynamic Semantic Graph Construction and Reasoning for Explainable Multi-hop Science Question Answering
• Answering Count Questions with Structured Answers from Text
• A Data-centric Framework for Improving Domain-specific Machine Reading Comprehension Datasets
• Question-Aware Memory Network for Multi-hop Question Answering in Human-Robot Interaction
• Ember: No-Code Context Enrichment via Similarity-Based Keyless Joins
• PALI at SemEval-2021 Task 2: Fine-Tune XLM-RoBERTa for Word in Context Disambiguation
• Sattiy at SemEval-2021 Task 9: An Ensemble Solution for Statement Verification and Evidence Finding with Tables
• RefBERT: Compressing BERT by Referencing to Pre-computed Representations