LayerRecall: A State-Conditioned Memory Router for Long-Horizon Consistency in Video Generation

  • 类型:arxiv
  • 标识:2608.28460
  • 链接:http://arxiv.org/abs/2608.28460v1
  • 主分类:multimodal
  • 形态:method
  • TLDR:Autoregressive video diffusion enables scalable long-video generation by producing chunks from a bounded recent context. While recency-based caching preserves local continuity, it evicts historical cues needed when subjects, objects, scenes, or attributes reappear. Existing memory mechanisms expose models to nonlocal history, but access alone does not ensure effective use. Our analysis reveals that video DiT layers exhibit distinct preferences for current, recent, and distant context, suggesting that long-range memory requires deciding both what to retrieve and where to use it. We introduce La
  • 待LLM分类:否
  • 来源文件
  • /inbox/tom/_candidates/2026-08-31-agent-rag-longcontext-candidates.json