Identifying Supporting Facts for Multi-hop Question Answering with Document Graph Networks

TL;DR

This paper introduces Document Graph Network (DGN), a message passing architecture that leverages graph-structured text representations to identify supporting facts in multi-hop question answering, improving reasoning capabilities.

Contribution

The paper presents DGN, a novel graph-based model for multi-hop reasoning that outperforms baseline models on HotpotQA by effectively utilizing structured text representations.

Findings

DGN achieves competitive results on HotpotQA.

Structured graph representations enhance multi-hop reasoning.

DGN outperforms baseline models operating on raw text.

Abstract

Recent advances in reading comprehension have resulted in models that surpass human performance when the answer is contained in a single, continuous passage of text. However, complex Question Answering (QA) typically requires multi-hop reasoning - i.e. the integration of supporting facts from different sources, to infer the correct answer. This paper proposes Document Graph Network (DGN), a message passing architecture for the identification of supporting facts over a graph-structured representation of text. The evaluation on HotpotQA shows that DGN obtains competitive results when compared to a reading comprehension baseline operating on raw text, confirming the relevance of structured representations for supporting multi-hop reasoning.