Graph-free Multi-hop Reading Comprehension: A Select-to-Guide Strategy
arXiv:2107.11823
Abstract
Multi-hop reading comprehension (MHRC) requires not only to predict the correct answer span in the given passage, but also to provide a chain of supporting evidences for reasoning interpretability. It is natural to model such a process into graph structure by understanding multi-hop reasoning as jumping over entity nodes, which has made graph modelling dominant on this task. Recently, there have been dissenting voices about whether graph modelling is indispensable due to the inconvenience of the graph building, however existing state-of-the-art graph-free attempts suffer from huge performance gap compared to graph-based ones. This work presents a novel graph-free alternative which firstly outperform all graph models on MHRC. In detail, we exploit a select-to-guide (S2G) strategy to accurately retrieve evidence paragraphs in a coarse-to-fine manner, incorporated with two novel attention mechanisms, which surprisingly shows conforming to the nature of multi-hop reasoning. Our graph-free model achieves significant and consistent performance gain over strong baselines and the current new state-of-the-art on the MHRC benchmark, HotpotQA, among all the published works.
References in corpus (10)
- Teaching Machines to Read and Comprehend
- Longformer: The Long-Document Transformer
- Bidirectional Attention Flow for Machine Comprehension
- Big Bird: Transformers for Longer Sequences
- Exploring Graph-structured Passage Representation for Multi-hop Reading Comprehension with Graph Neural Networks
- Multi-hop Question Answering via Reasoning Chains
- Is Graph Structure Necessary for Multi-hop Question Answering?
- Filling the Gap of Utterance-aware and Speaker-aware Representation for Multi-turn Dialogue
- Multi-span Style Extraction for Generative Reading Comprehension
- Learning Dense Representations of Phrases at Scale