{
  "id": 16554904,
  "title": "Closing the Modality Gap via Simplex-Constrained Representations",
  "url": "https://urgent.news/2026/10/23/closing-the-modality-gap-via-simplex-constrained-representations",
  "topic": "ai",
  "section": "AI",
  "published": "2026-10-23T16:57:48.000Z",
  "source": {
    "name": "arXiv cs.AI",
    "slug": "arxiv-cs-ai",
    "url": "https://arxiv.org/abs/2610.32890v1"
  },
  "original_language": "en",
  "account": null,
  "summary": "While multimodal contrastive models like CLIP and CLAP successfully learn shared representations, their embeddings typically exhibit a persistent \\emph{modality gap}. We propose addressing this separation through a purely geometric approach: changing the embedding domain from the unit sphere, $S^{d-1}$, to the probability simplex, $Δ^{d-1}$. Although both manifolds share the same intrinsic…",
  "key_points": [],
  "editors_take": null,
  "illustration": null,
  "coverage": {
    "outlets": 1,
    "also_reported_by": []
  },
  "ai_generated": true,
  "disclaimer": "Summaries, key points and the editor’s take are written by software from other outlets’ reporting and may contain errors — always check the linked original."
}