{
  "id": 12267,
  "url": "https://arxiv.org/abs/2607.17625v1",
  "title": "Brain-Aligned Multi-Stream Video Transformers with Sparse Self-Selection",
  "summary": "Modern video transformers typically ignore principles from primate vision and are rarely evaluated against neural data, limiting their biological interpretability. We introduce a sparse winner-takes-all token selection module that replaces dense self-attention to improve efficiency and approximate competitive routing observed in biological visual circuits. We further propose a neuro-inspired split-and-fuse video transformer which uses two complementary pathways: a high-resolution, low-frame-rate",
  "authors": "Amir Hosein Fadaei, Mahyar Maleki, Mohammad-Reza A. Dehaqani",
  "category": "research",
  "topics": "safety-alignment,biotech",
  "orgs": null,
  "regions": null,
  "published_at": "2026-07-20T07:30:16.000Z",
  "fetched_at": "2026-07-21T05:10:12.656Z",
  "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/12267",
  "original_url": "https://arxiv.org/abs/2607.17625v1",
  "evidence_status": "source-only",
  "attribution": "via ethics.ai"
}