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

2025-02-07

AI Summary

This study addresses the problem of predicting how a single utterance dynamically alters interpersonal impressions in speed-dating interactions, conditioned on both speaker and listener personality traits. We propose the first utterance-level, personality-aware impression change prediction framework: it leverages a large language model to generate candidate responses, injects bidirectional Big Five personality embeddings (for both interlocutors), and employs a multi-layer perceptron classifier to predict whether an utterance improves the listener’s impression. Human evaluation validates the naturalness, attractiveness, and impression-enhancement efficacy of generated dialogues. Our key contribution is the first integration of bidirectional personality modeling into fine-grained, utterance-level impression evolution prediction—overcoming the limitation of prior work that neglects dynamic personality–impression couplings. Experiments demonstrate that incorporating personality features yields an 8.2% absolute improvement in prediction accuracy; human evaluation further confirms that our generated utterances significantly outperform baselines in impression improvement, naturalness, and attractiveness.

Technology Category

• Natural Language Processing: Generation
• Machine Learning: Large Multimodal Models (LMMs)
• Humans and AI: Human-Aware Planning and Behavior Prediction

Application Category

• User Modeling, Personalization and Recommendation: Large Language Models (LLM) for user modeling and recommendation
• Social Networks and Social Media: Generative AI / large language models and their impact on social systems
• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions

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.

Problem

• Predicting impression change in speed dating
• Selecting utterances based on personalities
• Improving dialogue simulation effectiveness

Innovation

• Personality-based utterance selection
• Impression change prediction
• Simulated speed dating dialogues