跳到正文
arXiv cs.AI· Paul Bouchaud, Pedro Ramaciotti·· 10 天前AI 评分42

X 推荐系统无意中刻画用户意识形态立场

Recommender system in X inadvertently profiles ideological positions of users

AI 导读

研究基于 X 向法国 682 名志愿者展示的 250 万条「Who to Follow」推荐,重建了 X 推荐系统对 26,509 个账号的嵌入向量近似值,并计算经调查校准的意识形态分数。该嵌入中存在一个按左—右排序的方向(Pearson rho = 0.887),与年龄、性别或热度方向不同,且会影响推荐结果;移除该方向可在准确率小幅下降的情况下提升推荐多样性。

正文 · 原文

View PDF HTML (experimental)

Abstract:Several data protection laws restrict processing that reveals political opinions, irrespective of the controller's intent. Whether recommender systems do so as a by-product of optimizing relevance has not been measured. From 2.5 million ``Who to Follow'' recommendations shown to 682 volunteers in France, we reconstructed an approximation of the embedding used by X's recommender for 26,509 accounts, computing survey-calibrated ideology scores. One direction in this embedding orders users by Left-Right position (Pearson rho = 0.887), distinct from directions tracking age, gender or popularity. We show this scale exists and affects the recommendations computed from the embedding. Removing it diversified recommendations at a limited cost in accuracy. We document a consequential form of emergent political representation that current definitions of profiling do not clearly address.
Subjects: Social and Information Networks (cs.SI); Artificial Intelligence (cs.AI); Computers and Society (cs.CY)
Cite as: arXiv:2602.02624 [cs.SI]
  (or arXiv:2602.02624v2 [cs.SI] for this version)
  https://doi.org/10.48550/arXiv.2602.02624

arXiv-issued DOI via DataCite

Submission history

From: Pedro Ramaciotti [view email]
[v1] Mon, 2 Feb 2026 16:22:56 UTC (3,717 KB)
[v2] Tue, 6 Oct 2026 12:04:17 UTC (10,250 KB)

来源:arXiv cs.AI · arxiv.org