{
  "id": 10846328,
  "title": "Breaking the Uniformity Trap: Scaling Video Diffusion Model via SplitMoE",
  "url": "https://urgent.news/2026/09/29/breaking-the-uniformity-trap-scaling-video-diffusion-model-via",
  "topic": "ai",
  "section": "AI",
  "published": "2026-09-29T17:55:19.000Z",
  "source": {
    "name": "arXiv cs.AI",
    "slug": "arxiv-cs-ai",
    "url": "https://arxiv.org/abs/2609.38140v1"
  },
  "original_language": "en",
  "account": null,
  "summary": "Mixture-of-Experts (MoE), popularized by large language models, is a promising paradigm for scaling visual generative models. However, conventional token-wise MoE routes tokens independently within a homogeneous expert pool and regularizes expert usage toward uniformity, making it poorly matched to video data that is spatiotemporally redundant and semantically long-tailed. We show that existing…",
  "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."
}