研究人员开发了一个新的框架来审计大型语言模型（LLM）在在线讨论中模拟用户立场的方式。该框架测试了LLM模拟对对话上下文变化的敏感性，包括模因等模态元素。研究发现，LLM可以根据修订后的上下文有效地改变模拟立场，这既突显了使用这些模型模仿在线意见动态的潜力，也揭示了其风险。 AI
影响 凸显了LLM准确建模或操纵在线意见动态的潜力，强调了进行健全审计的必要性。
排序理由 该集群包含一篇学术论文，详细介绍了LLM立场模拟的新评估框架。
AI 生成摘要 · Google Gemini · 来自 3 个来源。 我们如何撰写摘要 →
arXiv:2606.06443v1 Announce Type: new 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 w…
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…
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.