Prototype-to-Style: Dialogue Generation with Style-Aware Editing on Retrieval Memory

TL;DR

This paper presents a prototype-to-style framework for stylistic dialogue generation that leverages retrieval memory and style-aware learning to produce high-quality, style-consistent responses, improving over existing methods.

Contribution

The paper introduces a novel prototype-to-style framework with a style-aware learning objective and de-noising strategy for improved stylistic dialogue generation.

Findings

Significant performance improvements over baselines

Effective cross-domain generalization

High-quality stylistic responses in multiple languages

Abstract

The ability of a dialog system to express prespecified language style during conversations has a direct, positive impact on its usability and on user satisfaction. We introduce a new prototype-to-style (PS) framework to tackle the challenge of stylistic dialogue generation. The framework uses an Information Retrieval (IR) system and extracts a response prototype from the retrieved response. A stylistic response generator then takes the prototype and the desired language style as model input to obtain a high-quality and stylistic response. To effectively train the proposed model, we propose a new style-aware learning objective as well as a de-noising learning strategy. Results on three benchmark datasets from two languages demonstrate that the proposed approach significantly outperforms existing baselines in both in-domain and cross-domain evaluations