Pretrained Persona Mixture Models and Tandem Models for Human Simulation

2026-09-18

AI Summary

本文针对当前大语言模型在模拟人类时存在的刻板预测问题，提出使用预训练的Persona混合模型和串联模型方法来提高模拟准确性和多样性。

Abstract

We argue here that the current dominant practice in LLM human simulation: prompting instruction-tuned assistant language models to role-play personas, is inaccurate and produces stereotyped predictions (lacking natural diversity). It has previously been shown that LLMs can be bound to personas using naturalistic, freetext dialog avoiding stereotyping. Here we show that binding can also be achieved using short, individual samples of dialog from specific people. Demographics can be added later without negative effects by simply querying the model. We use the term Persona Mixture Models (PMMs) for well-calibrated human models, currently realized as pretrained base models. We show that PMMs produce more accurate predictions than instruction-tuned models and retain more of the lexical, semantic, and pragmatic diversity found in human dialog. We measure realism and diversity of LLMs simulating human interlocutors across a diverse set of corpora spanning open-domain text, human-AI chat, and task-oriented dialogue between human speakers. However, base pretrained models can produce out-of-domain dialog and may lose some of the human's internal state over long contexts. We propose and explore tandem models which combine a pre-trained model with an instruction-tuned supervisor. Tandem models achieve the best overall accuracy and diversity in our experiments.

Problem

• Human Simulation
• Large Language Models
• Persona Mixture Models
• Stereotyped Predictions
• Diversity

Innovation

• Persona Mixture Models
• tandem models
• instruction-tuned