修正語境，轉變模擬立場：審計基於LLM的線上討論立場模擬

摘要

大型語言模型日益被用來模擬社交媒體使用者，並推斷個人可能如何在線上對話中回應。然而，這些模擬是否反映精確的個人信念，或者是否對對話情境中語義獨立的變化高度敏感，目前仍不明確。在本研究中，我們探討反事實情境修改作為審計基於大型語言模型的立場模擬之框架。對於一場原始線上對話，我們首先推斷目標使用者針對特定主題的立場。接著，我們對對話情境施以受控的修改策略，並在修改後的情境下再次模擬使用者的立場。我們比較純文字修改策略與納入迷因式情境的多模態策略，並評估兩個主要有效性指標，即平均方向性立場轉移與立場轉換率。結果顯示，在不同的極化偏好機制下，純文字與多模態策略均能實現有效且穩健的立場轉換。本研究提出一個評估框架，用以理解基於大型語言模型的立場模擬之情境敏感性。更廣泛而言，本研究凸顯了使用大型語言模型模擬線上意見動態的潛力與風險。

English

Large language models are increasingly used to simulate social media users and infer how individuals may respond to online discussions. However, it remains unclear whether these simulations reflect precise user-specific beliefs or whether they are highly sensitive to semantically independent changes in conversational contexts. In this work, we study counterfactual context revision as a framework for auditing LLM-based stance simulation. Given an original online conversation, we first infer a target user's stance toward a specific topic. We then apply controlled revision strategies to the conversational context and simulate the user's stance again under the revised context. We compare text-only revision strategies with a multimodal one that incorporates meme-based context and evaluate two main effectiveness metrics, i.e., average directional stance shift and stance transition rate. The results reveal effective and robust stance transitions in both text-only and multimodal strategies across different polarization-preference mechanisms. Our study contributes an evaluation framework for understanding the context sensitivity of LLM-based stance simulation. More broadly, it highlights both the promise and risk of using LLMs to simulate online opinion dynamics.