Answering Complex Open-Domain Questions with Multi-Hop Dense Retrieval
arXiv:2009.12756
Abstract
We propose a simple and efficient multi-hop dense retrieval approach for answering complex open-domain questions, which achieves state-of-the-art performance on two multi-hop datasets, HotpotQA and multi-evidence FEVER. Contrary to previous work, our method does not require access to any corpus-specific information, such as inter-document hyperlinks or human-annotated entity markers, and can be applied to any unstructured text corpus. Our system also yields a much better efficiency-accuracy trade-off, matching the best published accuracy on HotpotQA while being 10 times faster at inference time.
References in corpus (13)
- Exploring the Limits of Transfer Learning with a Unified Text-to-Text Transformer
- HuggingFace's Transformers: State-of-the-art Natural Language Processing
- Retrieval-Augmented Generation for Knowledge-Intensive NLP Tasks
- Longformer: The Long-Document Transformer
- REALM: Retrieval-Augmented Language Model Pre-Training
- End-to-End Open-Domain Question Answering with BERTserini
- Variational Reasoning for Question Answering with Knowledge Graph
- Coarse-grain Fine-grain Coattention Network for Multi-evidence Question Answering
- Differentiable Reasoning over a Virtual Knowledge Base
- PullNet: Open Domain Question Answering with Iterative Retrieval on Knowledge Bases and Text
- Hierarchical Graph Network for Multi-hop Question Answering
- Multi-hop Reading Comprehension across Multiple Documents by Reasoning over Heterogeneous Graphs
- UKP-Athene: Multi-Sentence Textual Entailment for Claim Verification
Cited by in corpus (6)
- Approximate Nearest Neighbor Negative Contrastive Learning for Dense Text Retrieval
- Retrieving and Reading: A Comprehensive Survey on Open-domain Question Answering
- A Graph-guided Multi-round Retrieval Method for Conversational Open-domain Question Answering
- Baleen: Robust Multi-Hop Reasoning at Scale via Condensed Retrieval
- More Robust Dense Retrieval with Contrastive Dual Learning
- HELIOS: Harmonizing Early Fusion, Late Fusion, and LLM Reasoning for Multi-Granular Table-Text Retrieval