Enhancing Impression Change Prediction in Speed Dating Simulations Based on Speakers' Personalities

TL;DR

The paper tackles the problem of predicting whether a target utterance in a speed-dating dialogue improves the partner's impression, conditioned on both speakers' personalities. It introduces a transformer-based classifier trained on a richly annotated MMSD corpus, formalizing the task as $T = f(P,u,D)$ and demonstrating that incorporating both personality and dialogue context enhances per-utterance impression prediction. Through a human-evaluation study, the authors show that using personality-aware utterance selection can produce dialogues that are more favorably received, compared to baselines that ignore per-utterance impression changes. The work highlights the practical potential of personality-informed dialogue generation for more efficient real-world speed-dating interactions, while noting data and ethical considerations as avenues for future work.

Abstract

This paper focuses on simulating text dialogues in which impressions between speakers improve during speed dating. This simulation involves selecting an utterance from multiple candidates generated by a text generation model that replicates a specific speaker's utterances, aiming to improve the impression of the speaker. Accurately selecting an utterance that improves the impression is crucial for the simulation. We believe that whether an utterance improves a dialogue partner's impression of the speaker may depend on the personalities of both parties. However, recent methods for utterance selection do not consider the impression per utterance or the personalities. To address this, we propose a method that predicts whether an utterance improves a partner's impression of the speaker, considering the personalities. The evaluation results showed that personalities are useful in predicting impression changes per utterance. Furthermore, we conducted a human evaluation of simulated dialogues using our method. The results showed that it could simulate dialogues more favorably received than those selected without considering personalities.

Figures (1)

• Figure 1: Example of expected dialogue simulation.