Jun 2026· arXiv.org· Vol abs/2606.06443· 0 citations· 36 references
Computer Science
TL;DR
This work studies counterfactual context revision as a framework for auditing LLM-based stance simulation and reveals effective and robust stance transitions in both text-only and multimodal strategies across different polarization-preference mechanisms.
Abstract
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 (https://github.com/STARResearchLab/StanceShift). More broadly, it highlights both the promise and risk of using LLMs as proxies for social media users in social simulation.
Understanding the dynamics of stance change on social media is crucial for addressing polarization and information integrity, yet observational studies face challenges including limited experimental control, restricted data access, and algorithmic confounds. We leverage Generative Agent-Based Modeling (GABM)—a novel si...
Valerio La Gatta, Gian Marco Orlando, Marco Perillo et al.· Proceedings of the 37th ACM...· 1 citation
Experimental results show that LLMs exhibit social bias even in coherent, response-conditioned multi-turn interactions, revealing late-emerging bias, non-monotonic bias patterns, and bias re-emergence, which motivate evaluations that extend beyond fixed-turn, pre-scripted protocols.
Rem Hida, Masahiro Kaneko, Daisuke Oba et al.· 1 citation
The results point to a fidelity gap in AI-mediated communication, with implications for journalism, policy deliberation, scientific communication, and other domains where opinion-laden messages pass through language models.
Ling-Chong Liu, Yan-Fei Zhou, Jacob Bien et al.· 0 citations
A framework that disentangles two distinct triggers of political sycophancy: opinion (aligning with explicit narratives) and identity (stereotyping based on demographic labels) is introduced, highlighting how personalization may amplify identity- or opinion-conditioned shifts in the model's behaviors.
Li-Ni Fu, Chang-Chih Meng, Chien-Hua Chen et al.· 0 citations
Large language models (LLMs) are increasingly used to answer questions about politically contentious issues, yet evaluations typically treat a model's stance as a relatively stable property. Real users, however, communicate political signals through their terminology, assumptions, and personal context. We investigate w...
Olivia Macmillan-Scott, Michael B. Jacobs, Nils W. Metternich et al.· arXiv.org· 0 citations