Listener's Social Identity Matters in Personalised Response Generation

TL;DR

This paper investigates how considering the listener's social identity, specifically gender, improves personalised response generation in Chinese social media dialogues, leading to more accurate and identity-aware responses.

Contribution

It introduces a method to incorporate the listener's social identity into personalised dialogue response generation, demonstrating improved performance and identity modeling.

Findings

Listener's identity influences language use in responses.

Modeling listener's identity improves generator performance.

Response generator captures social identity differences.

Abstract

Personalised response generation enables generating human-like responses by means of assigning the generator a social identity. However, pragmatics theory suggests that human beings adjust the way of speaking based on not only who they are but also whom they are talking to. In other words, when modelling personalised dialogues, it might be favourable if we also take the listener's social identity into consideration. To validate this idea, we use gender as a typical example of a social variable to investigate how the listener's identity influences the language used in Chinese dialogues on social media. Also, we build personalised generators. The experiment results demonstrate that the listener's identity indeed matters in the language use of responses and that the response generator can capture such differences in language use. More interestingly, by additionally modelling the listener's identity, the personalised response generator performs better in its own identity.