Exploring Graph-structured Passage Representation for Multi-hop Reading Comprehension with Graph Neural Networks

By Linfeng Song, Zhiguo Wang, Mo Yu, Radu Florian, Daniel Gildea. 2018

112 citations recorded by the source corpus.

Multi-hop reading comprehension focuses on one type of factoid question, where a system needs to properly integrate multiple pieces of evidence to correctly answer a question. Previous work approximates global evidence with local coreference information, encoding coreference chains with DAG-styled GRU layers within a gated-attention reader. However, coreference is limited in providing information for rich inference. We introduce a new method for better connecting global evidence, which forms more complex graphs compared to DAGs. To perform evidence integration on our graphs, we investigate two recent graph neural networks, namely graph convolutional network (GCN) and graph recurrent network (GRN). Experiments on two standard datasets show that richer global information leads to better answers. Our method performs better than all published results on these datasets.

Subfields: Artificial Intelligence, Natural Language Processing, Graph Learning

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Natural Language Processing

In Graph Learning

Citations in the corpus

Knowledge Guided Text Retrieval and Reading for Open Domain Question Answering (2019, background): “Most work concatenates passages into a single sequence (Swayamdipta et al., 2018; Yang et al., 2018; Song et al., 2018) or reads each in parallel (Clark and Gardner, 2018; Alberti et al., 2019; Min et al., 2019b; Wang et al., 2019).”
Multi-passage BERT: A Globally Normalized BERT Model for Open-domain Question Answering (2019, background): “In future, we plan to consider inter-correlation among passages for open-domain question answering (Wang et al., 2018b; Song et al., 2018).”
Hierarchical Graph Network for Multi-hop Question Answering (2019, methodology): “For example, MHQA-GRN (Song et al., 2018) and Coref-GRN (Dhingra et al., 2018) construct an entity graph based on co-reference resolution or sliding windows.”
Multi-hop Question Answering via Reasoning Chains (2019, methodology): “More recent work including Entity-GCN (De Cao et al., 2018), MHQA-GRN (Song et al., 2018), and BAG (Cao et al., 2019), form this problem as a search over entity graph, and adapt graph convolution network (Kipf and Welling, 2017) to do reasoning.”
Reasoning with Latent Structure Refinement for Document-Level Relation Extraction (2020, methodology): “Song et al. (2018a) and (De Cao et al., 2019) leverage co-reference information and set of rules to construct document-level entity graph.”
Do Multi-hop Readers Dream of Reasoning Chains? (2019, unknown): “More recent development of QA systems (Song et al., 2018; De Cao et al., 2018; Zhong et al., 2019) has started to focus on multi-hop reasoning on text passages, aiming to propose more sophisticated models beyond the shallow matching between questions and answers.”
Is Graph Structure Necessary for Multi-hop Question Answering? (2020, background): “Dhin-gra et al. (2018) and Song et al. (2018) designed a DAG-styled recurrent layer to model the relations between entities.”
Explore, Propose, and Assemble: An Interpretable Model for Multi-Hop Reading Comprehension (2019, methodology): “Song et al. (2018a) and De Cao et al. (2018) utilized Graph Convolutional Networks (Kipf and Welling, 2017) and Graph Recurrent Networks (Song et al., 2018b; Zhang et al., 2018) to model the relations between entities.”
Extracting Multiple-Relations in One-Pass with Pre-Trained Transformers (2019, background): “, entity-centric passage encoding in question answering (Song et al., 2018a).”
Understanding Dataset Design Choices for Multi-hop Reasoning (2019, background): “, 2018), MHQAGRN (Song et al., 2018), and Entity-GCN (De Cao et al.”
Dynamically Fused Graph Network for Multi-hop Reasoning (2019, methodology): “…network (Kipf and Welling, 2017), graph attention network (Veliˇckovi´c et al., 2018), and graph recurrent network (Song et al., 2018b), have been previously studied and show promising results in QA tasks requiring reasoning (Dhingra et al., 2018; De Cao et al., 2018; Song et al., 2018a).”
Multi-Step Inference for Reasoning over Paragraphs (2020, background): “, 2019) and WIKIHOP (Welbl et al., 2018; Song et al., 2018; Das et al., 2019; Asai et al., 2019) which aims to get the answer across the documents.”
Select, Answer and Explain: Interpretable Multi-hop Reading Comprehension over Multiple Documents (2019, methodology): “The other direction resorts to Graph Neural Networks (GNN) to realize multi-hop reasoning across multiple documents, and promising performance has been achieved (Song et al. 2018; De Cao, Aziz, and Titov 2019; Cao, Fang, and Tao 2019; Tu et al. 2019; Xiao et al. 2019).”
Repurposing Entailment for Multi-Hop Question Answering Tasks (2019, background): “Most approaches on these tasks perform simple attention-based aggregation (Mihaylov et al., 2018; Song et al., 2018; Cao et al., 2018) and do not exploit the entailment models trained on large scale datasets.”
Deep Learning with Graph Convolutional Networks: An Overview and Latest Applications in Computational Intelligence (2023, methodology): “[109] applied graph convolutional neural networks to reading comprehension, and the abstract meaning graph is applied to tasks such as text generation and relation extraction.”
Is Graph Structure Necessary for Multi-hop Reasoning? (2020, unknown): “Dhingra et al. (2018) and Song et al. (2018) designed a DAG-styled recurrent layer to model the relations between entities.”