Prototype-to-Style: Dialogue Generation with Style-Aware Editing on Retrieval Memory
This work addresses the problem of enhancing user satisfaction in dialogue systems by enabling style-specific responses, representing an incremental advancement in stylistic dialogue generation.
The paper tackles the challenge of generating dialogues with specific language styles by introducing a prototype-to-style framework that uses retrieval and style-aware editing, achieving significant performance improvements over existing baselines on three benchmark datasets across two languages.
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