批评Zak · 2025-08
“the authors have not convinced me that they will behave like people in this context.”
用 LLM 智能体模拟社交平台,测试六种改善干预的效果
论文把大语言模型嵌入基于智能体的模型(ABM),搭建只有发帖、转发和关注的极简社交平台,发现其中自发出现党派回音室、少数精英集中影响力、极化声音被放大三种已知弊病。作者测试了从按时间排序到桥接算法在内的六种干预,只看到有限改善,有时甚至更糟,推测问题根源在于互动反馈与网络增长之间的动态。
评论者引用它说明极化很难靠常见干预修复,甚至所有干预都无效或更糟;也有人质疑 LLM 智能体能否代表真实的人。
“the authors have not convinced me that they will behave like people in this context.”
“A recent study[1] seems to indicate that polarization is a hard problem, along with some of the other negative effects of social media.”
“I read a paper about social media being the way it is because of it's scale above other factors[0].”
“Notably, all attempts at conscious intervention failed to help or made things worse.”