{
  "id": 4606,
  "url": "https://arxiv.org/abs/2605.11033v1",
  "title": "TokaMind for Power Grid: Cross-Domain Transfer from Fusion Plasma",
  "summary": "TokaMind is a multi-modal transformer (MMT) foundation model pre-trained on tokamak plasma diagnostics data from MAST, where it was shown to outperform CNN-based approaches on fusion benchmarks. We investigate whether its learned representations generalize to physically distinct but structurally analogous domains. Through systematic experimentation across four domains-industrial bearing degradation, NASA CMAPSS turbofan degradation, and two independent power grid PMU datasets-we identify four tr",
  "authors": "JC Wu, Norton Lee, Kai Siang Chen",
  "category": "research",
  "topics": "healthcare,finance-investment",
  "orgs": null,
  "regions": null,
  "published_at": "2026-05-10T23:38:26.000Z",
  "fetched_at": "2026-07-14T16:31:08.355Z",
  "source_slug": "arxiv-ethics",
  "source_name": "arXiv",
  "source_homepage": "https://arxiv.org",
  "ethics_ai_record_url": "https://ethics.ai/record/4606",
  "original_url": "https://arxiv.org/abs/2605.11033v1",
  "evidence_status": "source-only",
  "attribution": "via ethics.ai"
}