Graph-based Hierarchical Relevance Matching Signals for Ad-hoc Retrieval

TL;DR

This paper introduces GHRM, a graph-based hierarchical relevance matching model that captures long-distance document relationships and hierarchical signals, improving ad-hoc retrieval effectiveness over existing methods.

Contribution

It proposes a novel GHRM model that explicitly models document-level word relationships with graph neural networks and captures hierarchical matching signals.

Findings

GHRM outperforms state-of-the-art methods on benchmark datasets.

Explicit graph modeling improves long-distance relationship capture.

Hierarchical signals enhance retrieval accuracy.

Abstract

The ad-hoc retrieval task is to rank related documents given a query and a document collection. A series of deep learning based approaches have been proposed to solve such problem and gained lots of attention. However, we argue that they are inherently based on local word sequences, ignoring the subtle long-distance document-level word relationships. To solve the problem, we explicitly model the document-level word relationship through the graph structure, capturing the subtle information via graph neural networks. In addition, due to the complexity and scale of the document collections, it is considerable to explore the different grain-sized hierarchical matching signals at a more general level. Therefore, we propose a Graph-based Hierarchical Relevance Matching model (GHRM) for ad-hoc retrieval, by which we can capture the subtle and general hierarchical matching signals simultaneously. We validate the effects of GHRM over two representative ad-hoc retrieval benchmarks, the comprehensive experiments and results demonstrate its superiority over state-of-the-art methods.