# 审计基于LLM的在线讨论立场模拟：反事实语境修正框架

- 来源：HuggingFace Daily Papers（社区热门论文）
- 发布时间：2026-06-04 08:00
- AIHOT 分数：52
- AIHOT 链接：https://aihot.virxact.com/items/cmq0zah490az8sltrl0att4x2
- 原文链接：https://arxiv.org/abs/2606.06443

## AI 摘要

本研究提出反事实语境修正框架，用于审计LLM在模拟社交媒体用户立场时的语境敏感性。给定原始对话后，先推断目标用户立场，再对语境施加受控修正策略（纯文本与结合模因的多模态策略）并重新模拟。评估平均方向性立场转变与立场转换率，发现两种策略在不同极化偏好机制下均实现有效且稳健的立场转换。该框架揭示了LLM立场模拟的语境敏感性，同时突出了其模拟在线舆论动态的前景与风险。

## 正文

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.