{
  "id": 17819,
  "url": "https://arxiv.org/abs/2608.06908v1",
  "title": "Calibrating WEAT Against Anisotropy: ZCA Whitening as a Geometric Pre-Processing Step for Embedding Association Tests",
  "summary": "We propose Zero-phase Component Analysis (ZCA) whitening as a geometric pre-processing step for the Word Embedding Association Test (WEAT). WEAT is a bias measurement method widely used in both computational social science and AI fairness research. It relies on cosine similarity as a measure of semantic association, which assumes that the embedding space is approximately isotropic. However, prior work has reported that many widely used language models do not satisfy this assumption, raising conc",
  "authors": "Seitaro Ono, Senna Ross, Jun Saiki",
  "category": "research",
  "topics": "bias-fairness",
  "orgs": null,
  "regions": null,
  "published_at": "2026-08-07T07:42:26.000Z",
  "fetched_at": "2026-08-10T05:10:00.488Z",
  "source_slug": "arxiv-ethics",
  "source_name": "arXiv",
  "source_homepage": "https://arxiv.org",
  "ethics_ai_record_url": "https://ethics.ai/record/17819",
  "original_url": "https://arxiv.org/abs/2608.06908v1",
  "evidence_status": "source-only",
  "attribution": "via ethics.ai"
}