{
  "id": 17396,
  "url": "https://arxiv.org/abs/2608.05315v1",
  "title": "Rectifying Geometric Misalignment: Online Source-Free Adaptation for Class-Imbalanced EEG",
  "summary": "Electroencephalography (EEG) based Brain-Computer Interfaces (BCIs) often require unsupervised domain adaptation (UDA) to generalize across subjects and sessions. While Riemannian alignment methods like the Riemannian Centering Transformation (RCT) are effective for handling covariate shifts, they implicitly assume balanced class priors. However, in realistic online BCI scenarios, the label distributions vary dynamically (label shift), causing standard alignment techniques to geometrically misal",
  "authors": "Shiwen Chu, Shanglin Li, Motoaki Kawanabe, Reinmar Kobler",
  "category": "research",
  "topics": "safety-alignment",
  "orgs": null,
  "regions": null,
  "published_at": "2026-08-05T18:17:59.000Z",
  "fetched_at": "2026-08-07T05:10:58.501Z",
  "source_slug": "arxiv-cslg",
  "source_name": "arXiv cs.LG",
  "source_homepage": "https://arxiv.org/list/cs.LG/recent",
  "ethics_ai_record_url": "https://ethics.ai/record/17396",
  "original_url": "https://arxiv.org/abs/2608.05315v1",
  "evidence_status": "source-only",
  "attribution": "via ethics.ai"
}