Conversational Query Rewriting with Self-supervised Learning

TL;DR

This paper introduces a self-supervised learning approach for conversational query rewriting, creating a large dataset and a Transformer-based model that improves multi-turn dialogue understanding without extensive manual annotation.

Contribution

The paper presents a novel self-supervised dataset construction method and a Transformer-based CQR model with keyword detection and intent constraints, advancing dialogue system capabilities.

Findings

Outperforms existing CQR baselines significantly

Self-supervised learning effectively improves CQR performance

Demonstrates robustness on multiple datasets

Abstract

Context modeling plays a critical role in building multi-turn dialogue systems. Conversational Query Rewriting (CQR) aims to simplify the multi-turn dialogue modeling into a single-turn problem by explicitly rewriting the conversational query into a self-contained utterance. However, existing approaches rely on massive supervised training data, which is labor-intensive to annotate. And the detection of the omitted important information from context can be further improved. Besides, intent consistency constraint between contextual query and rewritten query is also ignored. To tackle these issues, we first propose to construct a large-scale CQR dataset automatically via self-supervised learning, which does not need human annotation. Then we introduce a novel CQR model Teresa based on Transformer, which is enhanced by self-attentive keywords detection and intent consistency constraint. Finally, we conduct extensive experiments on two public datasets. Experimental results demonstrate that our proposed model outperforms existing CQR baselines significantly, and also prove the effectiveness of self-supervised learning on improving the CQR performance.