Large language models are increasingly used to simulate social media users and infer how individuals may respond to online discussions. In this work, we study counterfactual context revision as a framework for auditing LLM-based stance simulation.</p>\n","updatedAt":"2026-06-05T12:52:37.951Z","author":{"_id":"64c939307dba66c3a7e4d215","avatarUrl":"/avatars/b4c7f43b47db93ca5d7aa30e3d9ef80e.svg","fullname":"BruceLyu","name":"brucelyu","type":"user","isPro":false,"isHf":false,"isHfAdmin":false,"isMod":false,"followerCount":1,"isUserFollowing":false}},"numEdits":0,"identifiedLanguage":{"language":"en","probability":0.8598677515983582},"editors":["brucelyu"],"editorAvatarUrls":["/avatars/b4c7f43b47db93ca5d7aa30e3d9ef80e.svg"],"reactions":[],"isReport":false}}],"primaryEmailConfirmed":false,"paper":{"id":"2606.06443","authors":[{"_id":"6a22c60176dea4a01ef147cf","name":"Xinnong Zhang","hidden":false},{"_id":"6a22c60176dea4a01ef147d0","name":"Wanting Shan","hidden":false},{"_id":"6a22c60176dea4a01ef147d1","name":"Hanjia Lyu","hidden":false},{"_id":"6a22c60176dea4a01ef147d2","name":"Zhongyu Wei","hidden":false},{"_id":"6a22c60176dea4a01ef147d3","name":"Jiebo Luo","hidden":false}],"publishedAt":"2026-06-04T00:00:00.000Z","submittedOnDailyAt":"2026-06-05T00:00:00.000Z","title":"Revising Context, Shifting Simulated Stance: Auditing LLM-Based Stance Simulation in Online Discussions","submittedOnDailyBy":{"_id":"64c939307dba66c3a7e4d215","avatarUrl":"/avatars/b4c7f43b47db93ca5d7aa30e3d9ef80e.svg","isPro":false,"fullname":"BruceLyu","user":"brucelyu","type":"user","name":"brucelyu"},"summary":"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.","upvotes":2,"discussionId":"6a22c60276dea4a01ef147d4","ai_summary":"LLM-based stance simulation exhibits context sensitivity when subjected to counterfactual revisions, with both text-only and multimodal approaches showing robust stance transitions across different polarization mechanisms.","ai_keywords":["large language models","stance simulation","counterfactual context revision","conversational context","multimodal approach","polarization-preference mechanisms","stance transition rate"],"ai_summary_model":"Qwen/Qwen2.5-Coder-32B-Instruct"},"canReadDatabase":false,"canManagePapers":false,"canSubmit":false,"hasHfLevelAccess":false,"upvoted":false,"upvoters":[{"_id":"64c939307dba66c3a7e4d215","avatarUrl":"/avatars/b4c7f43b47db93ca5d7aa30e3d9ef80e.svg","isPro":false,"fullname":"BruceLyu","user":"brucelyu","type":"user"},{"_id":"6a20c8969629e903aa357ea1","avatarUrl":"/avatars/705453035c95ec78ade02d8f115ea301.svg","isPro":false,"fullname":"Shaohuan Zhang","user":"shaohuanzhang","type":"user"}],"acceptLanguages":["en"],"dailyPaperRank":0}">
Revising Context, Shifting Simulated Stance: Auditing LLM-Based Stance Simulation in Online Discussions
Abstract
LLM-based stance simulation exhibits context sensitivity when subjected to counterfactual revisions, with both text-only and multimodal approaches showing robust stance transitions across different polarization mechanisms.
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.
Community
Large language models are increasingly used to simulate social media users and infer how individuals may respond to online discussions. In this work, we study counterfactual context revision as a framework for auditing LLM-based stance simulation.
Upload images, audio, and videos by dragging in the text input, pasting, or clicking here.
Tap or paste here to upload images
Cite arxiv.org/abs/2606.06443 in a model README.md to link it from this page.
Cite arxiv.org/abs/2606.06443 in a dataset README.md to link it from this page.
Cite arxiv.org/abs/2606.06443 in a Space README.md to link it from this page.
Discussion (0)
Sign in to join the discussion. Free account, 30 seconds — email code or GitHub.
Sign in →No comments yet. Sign in and be the first to say something.