{
  "id": 12651,
  "url": "https://arxiv.org/abs/2607.20368",
  "title": "Self Gradient Forcing: Native Long Video Extrapolation",
  "summary": "Recent autoregressive video diffusion methods are increasingly built upon Self Forcing, where the student is trained on histories produced by its own rollout rather than ground-truth video contexts. This reduces exposure bias, but the historical key-value cache is still used by future frames only as frozen rollout state. As a result, future losses cannot supervise how earlier generated latents should be written into more useful keys and values for later video-latent generation. We call this the",
  "authors": "Junhao Zhuang, Shiyi Zhang, Yuxuan Bian, Yaowei Li, Yawen Luo, Yijun Liu",
  "category": "research",
  "topics": "bias-fairness,children-education",
  "orgs": null,
  "regions": null,
  "published_at": "2026-07-21T20:00:00.000Z",
  "fetched_at": "2026-07-23T05:10:49.458Z",
  "source_slug": "hf-daily",
  "source_name": "HuggingFace Daily Papers",
  "source_homepage": "https://huggingface.co/papers",
  "ethics_ai_record_url": "https://ethics.ai/record/12651",
  "original_url": "https://arxiv.org/abs/2607.20368",
  "evidence_status": "source-only",
  "attribution": "via ethics.ai"
}