Not logged in.

Contribution Details

Type Conference or Workshop Paper
Scope Discipline-based scholarship
Published in Proceedings Yes
Title QAGCN: Answering Multi-relation Questions via Single-Step Implicit Reasoning over Knowledge Graphs
Organization Unit
  • Ruijie Wang
  • Luca Rossetto
  • Michael Cochez
  • Abraham Bernstein
Presentation Type paper
Item Subtype Original Work
Refereed Yes
Status Published in final form
  • English
ISBN 9783031606250
ISSN 0302-9743
Page Range 41 - 58
Event Title The 21st Extended Semantic Web Conference (ESWC 2024)
Event Type conference
Event Location Hersonissos, Greece
Event Start Date May 26 - 2024
Event End Date May 30 - 2024
Series Name Lecture Notes in Computer Science
Number 14664
Publisher Springer
Abstract Text Multi-relation question answering (QA) is a challenging task, where given questions usually require long reasoning chains in KGs that consist of multiple relations. Recently, methods with explicit multi-step reasoning over KGs have been prominently used in this task and have demonstrated promising performance. Examples include methods that perform stepwise label propagation through KG triples and methods that navigate over KG triples based on reinforcement learning. A main weakness of these methods is that their reasoning mechanisms are usually complex and difficult to implement or train. In this paper, we argue that multi-relation QA can be achieved via end-to-end single-step implicit reasoning, which is simpler, more efficient, and easier to adopt. We propose QAGCN — a Question-Aware Graph Convolutional Network (GCN)-based method that includes a novel GCN architecture with controlled question-dependent message propagation for the implicit reasoning. Extensive experiments have been conducted, where QAGCN achieved competitive and even superior performance compared to state-of-the-art explicit-reasoning methods. Our code and pre-trained models are available in the repository:
Digital Object Identifier 10.1007/978-3-031-60626-7_3
Export BibTeX