Training Millions of Personalized Dialogue Agents

TL;DR

This paper introduces a large-scale dataset with 5 million personas and 700 million dialogues, demonstrating that persona-based training enhances end-to-end dialogue systems and achieves state-of-the-art results.

Contribution

The paper presents a new extensive dataset for personalized dialogue training and shows its effectiveness in improving dialogue system performance.

Findings

Persona conditioning improves engagement in dialogue models.

Large-scale datasets lead to better performance and generalization.

Fine-tuning on the new dataset achieves state-of-the-art results.

Abstract

Current dialogue systems are not very engaging for users, especially when trained end-to-end without relying on proactive reengaging scripted strategies. Zhang et al. (2018) showed that the engagement level of end-to-end dialogue models increases when conditioning them on text personas providing some personalized back-story to the model. However, the dataset used in Zhang et al. (2018) is synthetic and of limited size as it contains around 1k different personas. In this paper we introduce a new dataset providing 5 million personas and 700 million persona-based dialogues. Our experiments show that, at this scale, training using personas still improves the performance of end-to-end systems. In addition, we show that other tasks benefit from the wide coverage of our dataset by fine-tuning our model on the data from Zhang et al. (2018) and achieving state-of-the-art results.