Miracle: Towards Personalized Dialogue Generation with Latent-Space Multiple Personal Attribute Control
Zhenyi Lu, Wei Wei, Xiaoye Qu, Xian-Ling Mao, Dangyang Chen, Jixiong Chen
Abstract
Personalized dialogue systems aim to endow the chatbot agent with more anthropomorphic traits for human-like interactions. Previous approaches have explored explicitly user profile modeling using text descriptions, implicit derivation of user embeddings, or utilizing handicraft prompts for ChatGPT-like models. However, textual personas are limited in describing multi-faceted attributes (e.g., language style, inner character nuances), implicit embedding suffers from personality sparsity, and handicraft prompts lack fine-grained and stable controllability. Hence, these approaches may struggle with complex personalized dialogue generation tasks that require generating controllable responses with multiple personal attributes. To this end, we propose Miracle, a novel personalized dialogue generation method through MultIple PeRsonal Attributes Control within Latent-Space Energy-based Models. ttributes Control within Latent-Space Energy-based Models. Specifically, our approach first disentangles complex personality into multi-faceted attributes. Subsequently, we employ a conditional variational auto-encoder to align with the dense personalized responses within a latent joint attribute space. We have also tailored a dedicated energy function and customized the ordinary differential equations sampling method to offer flexible attribute composition and precise attribute control. Extensive experiments demonstrate that Miracle outperforms several strong baselines in terms of personality controllability and response generation quality. Our dataset and code are available at https://github.com/LZY-the-boys/MIRACLE- Anthology ID:
- 2023.findings-emnlp.395
- Volume:
- Findings of the Association for Computational Linguistics: EMNLP 2023
- Month:
- December
- Year:
- 2023
- Address:
- Singapore
- Editors:
- Houda Bouamor, Juan Pino, Kalika Bali
- Venue:
- Findings
- SIG:
- Publisher:
- Association for Computational Linguistics
- Note:
- Pages:
- 5933–5957
- Language:
- URL:
- https://aclanthology.org/2023.findings-emnlp.395
- DOI:
- 10.18653/v1/2023.findings-emnlp.395
- Cite (ACL):
- Zhenyi Lu, Wei Wei, Xiaoye Qu, Xian-Ling Mao, Dangyang Chen, and Jixiong Chen. 2023. Miracle: Towards Personalized Dialogue Generation with Latent-Space Multiple Personal Attribute Control. In Findings of the Association for Computational Linguistics: EMNLP 2023, pages 5933–5957, Singapore. Association for Computational Linguistics.
- Cite (Informal):
- Miracle: Towards Personalized Dialogue Generation with Latent-Space Multiple Personal Attribute Control (Lu et al., Findings 2023)
- PDF:
- https://preview.aclanthology.org/naacl24-info/2023.findings-emnlp.395.pdf