Learning to Retrieve Reasoning Paths over Wikipedia Graph for Question Answering

Abstract

Answering questions that require multi-hop reasoning at web-scale necessitates retrieving multiple evidence documents, one of which often has little lexical or semantic relationship to the question. This paper introduces a new graph-based recurrent retrieval approach that learns to retrieve reasoning paths over the Wikipedia graph to answer multi-hop open-domain questions. Our retriever model trains a recurrent neural network that learns to sequentially retrieve evidence paragraphs in the reasoning path by conditioning on the previously retrieved documents. Our reader model ranks the reasoning paths and extracts the answer span included in the best reasoning path. Experimental results show state-of-the-art results in three open-domain QA datasets, showcasing the effectiveness and robustness of our method. Notably, our method achieves significant improvement in HotpotQA, outperforming the previous best model by more than 14 points.

Published as a conference paper at ICLR 2020. Code is available at https://github.com/AkariAsai/learning_to_retrieve_reasoning_paths

References in corpus (12)

• Bidirectional Attention Flow for Machine Comprehension
• End-to-End Open-Domain Question Answering with BERTserini
• Passage Re-ranking with BERT
• HotpotQA: A Dataset for Diverse, Explainable Multi-hop Question Answering
• Dynamically Fused Graph Network for Multi-hop Reasoning
• PullNet: Open Domain Question Answering with Iterative Retrieval on Knowledge Bases and Text
• Cognitive Graph for Multi-Hop Reading Comprehension at Scale
• Efficient and Robust Question Answering from Minimal Context over Documents
• Real-Time Open-Domain Question Answering with Dense-Sparse Phrase Index
• Weaver: Deep Co-Encoding of Questions and Documents for Machine Reading
• Ranking Paragraphs for Improving Answer Recall in Open-Domain Question Answering
• Answering Complex Open-domain Questions Through Iterative Query Generation

Cited by in corpus (21)

• REALM: Retrieval-Augmented Language Model Pre-Training
• BARTScore: Evaluating Generated Text as Text Generation
• Subgraph Retrieval Enhanced Model for Multi-hop Knowledge Base Question Answering
• Multi-hop Question Answering via Reasoning Chains
• Generation-Augmented Retrieval for Open-domain Question Answering
• A Survey on Explainability in Machine Reading Comprehension
• Interpreting Deep Learning Models in Natural Language Processing: A Review
• MultiModalQA: Complex Question Answering over Text, Tables and Images
• Connecting the Dots: A Knowledgeable Path Generator for Commonsense Question Answering
• Multi-Step Inference for Reasoning Over Paragraphs
• Knowledge-Aided Open-Domain Question Answering
• Answering Questions on COVID-19 in Real-Time
• ReCo: Reliable Causal Chain Reasoning via Structural Causal Recurrent Neural Networks
• Dynamic Semantic Graph Construction and Reasoning for Explainable Multi-hop Science Question Answering
• SPARTA: Efficient Open-Domain Question Answering via Sparse Transformer Matching Retrieval
• Revisiting the Open-Domain Question Answering Pipeline
• Does Structure Matter? Leveraging Data-to-Text Generation for Answering Complex Information Needs
• Complementary Evidence Identification in Open-Domain Question Answering
• Learning to Recover Reasoning Chains for Multi-Hop Question Answering via Cooperative Games
• Interactive Fiction Game Playing as Multi-Paragraph Reading Comprehension with Reinforcement Learning
• Efficient Retrieval Optimized Multi-task Learning