Signals 4 · free daily AI digest

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

Paper recorded by Signals 4 on 2026-08-28 in cs.CV. Abstract reproduced from arXiv; link to the original below.

Published 2026-08-28 on arXiv · recorded by Signals 4 on 2026-08-31

Category: cs.CV · 计算机视觉 · first seen 2026-08-31

Abstract

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 tha

Read on arXiv →

#227 most recent of 237 cs.CV papers we have recorded · ↑ newer: Learning the Target Priors Before Image Translation: A Decoupled Train · ↓ older: Prompt-Guided Interactive Segmentation of Interstitial Lung Disease in
Cite this page: LayerRecall: A State-Conditioned Memory Router for Long-Horizon Consistency in Video Generation: the #227 most recent of 237 cs.CV papers we have recorded (as of 2026-08-28). Source: Signals 4 (Signals API) — https://data.jiangzhang.ca/signals4/t/papers/layerrecall-a-state-conditioned-memory-router-for-long-horizon-consistency-in-vi.html
Free to quote with attribution to “Signals 4 (Signals API)”. Machine-readable: papers.json
Related: More cs.CV papers · arXiv signals · All papers · Today in AI
Get 4 AI signals a day by email — free.
Subscribe free → See all plans →
Get 4 AI signals a day by email — free
All models · All repos · By company · Daily editions