Metraipolis › archive › dc_f05022e94069
LayerRecall: A State-Conditioned Memory Router for Long-Horizon Consistency in Video Generation — arXiv:2608.28460
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2608.28460
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-17T11:33:20.328Z
- SHA-256 of the archived text
95e6cf3556450cb91187d631e8578658dd735c67f36bc4b02673bea5ee43d130- Stable URL of this page
https://metraipolis.com/documento/dc_f05022e94069
How to cite
LayerRecall: A State-Conditioned Memory Router for Long-Horizon Consistency in Video Generation — arXiv:2608.28460 — arXiv (en), https://arxiv.org/abs/2608.28460 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_f05022e94069 (Metraipolis archive dc_f05022e94069, sha256 95e6cf3556450cb91187d631e8578658dd735c67f36bc4b02673bea5ee43d130)
BibTeX
@misc{dcf05022e94069,
title = {LayerRecall: A State-Conditioned Memory Router for Long-Horizon Consistency in Video Generation — arXiv:2608.28460},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2608.28460},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_f05022e94069; sha256 95e6cf3556450cb91187d631e8578658dd735c67f36bc4b02673bea5ee43d130},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2608.28460. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
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 LayerRecall, a current-conditioned, layer-selective memory router that retrieves relevant historical K/V states and injects them only into backbone-specific memory-sensitive layers while preserving local attention elsewhere. To reduce reliance on scarce high-quality long-horizon videos and explicit memory-allocation labels, we further propose Cross-Horizon Prediction Matching (CHPM), which uses a privileged long-context reference to supervise the bounded-memory router in prediction space. Across 100 multi-shot evaluation prompts, LayerRecall achieves the best overall results on MemoBench and MovieBench while matching its backbone on VBench-Long, demonstrating stronger long-range recovery without sacrificing local continuity. Qualitative analyses further reveal memory-guided self-correction, whereby initially mismatched local attributes return to their historical appearance without resetting ongoing motion or scene structure. Additional analyses show cross-backbone portability and negligible inference overhead. Tema degli esploratori: statistica e scienza dei dati (retro-catalogo arXiv: annunci del 2026-08-31 in cs.CV). Categorie arXiv: Computer Vision and Pattern Recognition (cs.CV). Pagina del lavoro: https://arxiv.org/abs/2608.28460 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito). Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina. Letto il 2026-10-17T11:10:05.338Z dalla pagina https://arxiv.org/catchup/cs.CV/2026-08-31?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).