DepWiGNN: A Depth-wise Graph Neural Network for Multi-hop Spatial Reasoning in Text

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Li, Shuaiyi, Deng, Yang, Lam, Wai
Format: Preprint
Published: 2023
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866913257439100928
author Li, Shuaiyi
Deng, Yang
Lam, Wai
author_facet Li, Shuaiyi
Deng, Yang
Lam, Wai
contents Spatial reasoning in text plays a crucial role in various real-world applications. Existing approaches for spatial reasoning typically infer spatial relations from pure text, which overlooks the gap between natural language and symbolic structures. Graph neural networks (GNNs) have showcased exceptional proficiency in inducing and aggregating symbolic structures. However, classical GNNs face challenges in handling multi-hop spatial reasoning due to the over-smoothing issue, i.e., the performance decreases substantially as the number of graph layers increases. To cope with these challenges, we propose a novel Depth-Wise Graph Neural Network (DepWiGNN). Specifically, we design a novel node memory scheme and aggregate the information over the depth dimension instead of the breadth dimension of the graph, which empowers the ability to collect long dependencies without stacking multiple layers. Experimental results on two challenging multi-hop spatial reasoning datasets show that DepWiGNN outperforms existing spatial reasoning methods. The comparisons with the other three GNNs further demonstrate its superiority in capturing long dependency in the graph.
format Preprint
id arxiv_https___arxiv_org_abs_2310_12557
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle DepWiGNN: A Depth-wise Graph Neural Network for Multi-hop Spatial Reasoning in Text
Li, Shuaiyi
Deng, Yang
Lam, Wai
Computation and Language
Artificial Intelligence
Spatial reasoning in text plays a crucial role in various real-world applications. Existing approaches for spatial reasoning typically infer spatial relations from pure text, which overlooks the gap between natural language and symbolic structures. Graph neural networks (GNNs) have showcased exceptional proficiency in inducing and aggregating symbolic structures. However, classical GNNs face challenges in handling multi-hop spatial reasoning due to the over-smoothing issue, i.e., the performance decreases substantially as the number of graph layers increases. To cope with these challenges, we propose a novel Depth-Wise Graph Neural Network (DepWiGNN). Specifically, we design a novel node memory scheme and aggregate the information over the depth dimension instead of the breadth dimension of the graph, which empowers the ability to collect long dependencies without stacking multiple layers. Experimental results on two challenging multi-hop spatial reasoning datasets show that DepWiGNN outperforms existing spatial reasoning methods. The comparisons with the other three GNNs further demonstrate its superiority in capturing long dependency in the graph.
title DepWiGNN: A Depth-wise Graph Neural Network for Multi-hop Spatial Reasoning in Text
topic Computation and Language
Artificial Intelligence
url https://arxiv.org/abs/2310.12557