MCP: Self-supervised Pre-training for Personalized Chatbots with Multi-level Contrastive Sampling

Abstract

Personalized chatbots focus on endowing the chatbots with a consistent personality to behave like real users and further act as personal assistants. Previous studies have explored generating implicit user profiles from the user's dialogue history for building personalized chatbots. However, these studies only use the response generation loss to train the entire model, thus it is prone to suffer from the problem of data sparsity. Besides, they overemphasize the final generated response's quality while ignoring the correlations and fusions between the user's dialogue history, leading to rough data representations and performance degradation. To tackle these problems, we propose a self-supervised learning framework MCP for capturing better representations from users' dialogue history for personalized chatbots. Specifically, we apply contrastive sampling methods to leverage the supervised signals hidden in user dialog history, and generate the pre-training samples for enhancing the model. We design three pre-training tasks based on three types of contrastive pairs from user dialogue history, namely response pairs, sequence augmentation pairs, and user pairs. We pre-train the utterance encoder and the history encoder towards the contrastive objectives and use these pre-trained encoders for generating user profiles while personalized response generation. Experimental results on two realworld datasets show a significant improvement in our proposed model MCP compared with the existing methods.

Miracle: Towards Personalized Dialogue Generation with Latent-Space Multiple Personal Attribute Control

“…As for embedding-based methods, traditional approaches (Li et al, 2016b;Al-Rfou et al, 2016) attempt to exploit user ID information, while DHAP (Ma et al, 2021) embed user dialogue history as implicit profiles. More recently, contrastive learning (Huang et al, 2022), refined retrieval (Zhong et al, 2022) and CVAE-based clustering (Tang et al, 2023) are explored to enhance the personalization performance. However, these approaches may still suffer from the personality scarcity of real-world posts without explicit modeling.…”

“…For instance, while a statement like "I grew up in the deep south" conveys traits related to regional identity, it overlooks other personality dimensions such as language style, attitudes, and inner character nuances. Other methods for personalized dialogue generation often rely on user embeddings derived from social media platforms like Reddit (Qian et al, 2021;Ma et al, 2021;Huang et al, 2022;Zhong et al, 2022). However, these models encounter challenges due to the sparsity present in real-world posts, as they lack explicit persona modeling.…”