A Document-grounded Matching Network for Response Selection in Retrieval-based Chatbots

TL;DR

This paper introduces a document-grounded matching network (DGMN) that enhances response selection in retrieval-based chatbots by effectively integrating background documents and conversation context, leading to improved accuracy and interpretability.

Contribution

The paper proposes a novel DGMN model that dynamically fuses document and context information for better response matching in knowledge-aware chatbots.

Findings

DGMN significantly outperforms state-of-the-art methods on public datasets.

The model demonstrates good interpretability in response selection.

Empirical results confirm the effectiveness of hierarchical interaction for grounding.

Abstract

We present a document-grounded matching network (DGMN) for response selection that can power a knowledge-aware retrieval-based chatbot system. The challenges of building such a model lie in how to ground conversation contexts with background documents and how to recognize important information in the documents for matching. To overcome the challenges, DGMN fuses information in a document and a context into representations of each other, and dynamically determines if grounding is necessary and importance of different parts of the document and the context through hierarchical interaction with a response at the matching step. Empirical studies on two public data sets indicate that DGMN can significantly improve upon state-of-the-art methods and at the same time enjoys good interpretability.