85137

Автор(ы): 

Автор(ов): 

2

Параметры публикации

Тип публикации: 

Доклад

Название: 

Modeling Opinion Formation in LLMs as a Social and Sequential Process

Наименование конференции: 

  • HCI International 2026 Conference (HCII 2026)

Наименование источника: 

  • Lecture Notes in Computer Science (LNCS)

Обозначение и номер тома: 

volume 16715

Город: 

  • Montreal, Canada

Издательство: 

  • Springer

Год издания: 

2026

Страницы: 

292-303
Аннотация
This study investigates a fundamental question in AI-human interaction: To what extent do language models internalize and propagate human opinions when forming their own judgments? Using business reviews as a microcosm of human opinion expression, we examine how LLMs sequentially update their evaluations based on human-generated content. Unlike traditional sentiment analysis that treats reviews independently, our framework models opinion formation as a social learning process where LLMs act as synthetic agents influenced by the expressed opinions of human reviewers. We reveal that LLMs not only analyze content but also calibrate to collective human sentiment, raising important questions about AI autonomy, social influence dynamics, and the construction of "artificial opinion" in review systems.

Библиографическая ссылка: 

Леонова Ю.С., Федянин Д.Н. Modeling Opinion Formation in LLMs as a Social and Sequential Process / Lecture Notes in Computer Science (LNCS). Montreal, Canada: Springer, 2026. volume 16715. С. 292-303.