Query Enhanced Knowledge-Intensive Conversation via Unsupervised Joint Modeling

TL;DR

This paper introduces QKConv, an unsupervised joint model for knowledge-intensive conversations that generates responses by exploring multiple queries and leveraging relevant knowledge without requiring extra annotations.

Contribution

The paper presents a novel unsupervised joint training approach for knowledge-intensive dialogue systems, eliminating the need for annotated queries or knowledge sources.

Findings

QKConv outperforms all unsupervised baselines on three datasets.

QKConv achieves competitive results with supervised methods.

Joint training effectively leverages dialogue context and target responses.

Abstract

In this paper, we propose an unsupervised query enhanced approach for knowledge-intensive conversations, namely QKConv. There are three modules in QKConv: a query generator, an off-the-shelf knowledge selector, and a response generator. QKConv is optimized through joint training, which produces the response by exploring multiple candidate queries and leveraging corresponding selected knowledge. The joint training solely relies on the dialogue context and target response, getting exempt from extra query annotations or knowledge provenances. To evaluate the effectiveness of the proposed QKConv, we conduct experiments on three representative knowledge-intensive conversation datasets: conversational question-answering, task-oriented dialogue, and knowledge-grounded conversation. Experimental results reveal that QKConv performs better than all unsupervised methods across three datasets and achieves competitive performance compared to supervised methods.