AI Paper Atlas

Research map / Video / Video generation

LUVE : Latent-Cascaded Ultra-High-Resolution Video Generation with Dual Frequency Experts

Chen Zhao, Jiawei Chen, Hongyu Li, Zhuoliang Kang, Shilin Lu, Xiaoming Wei, Kai Zhang, Jian Yang, Ying Tai

ICML 2026 · Video › Video generation · approach: Autoregressive long video generation · 0 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
replace RGB/latent interpolation with the learned video latent upsampler (VLUer) (ablation)LUVE 2K · 2K ablationFID_patch: 51.75 (RGB) / 47.80 (latent) -> 41.03Improved
replace RGB interpolation with latent-space upsampler (ablation)LUVE 2K · 2K ablationlatency: 40.12s -> 0.922sImproved
use learned upsampler instead of plain latent interpolation (ablation)LUVE 2K · 2K ablationlatency: 0.004s -> 0.922sHurt
add cascaded low-res motion stage over end-to-end UHR scaling (ablation)LUVE 2K · 2K ablationFID_patch: 54.10 -> 46.48Improved
add two standard LoRA experts (not frequency-specific) (ablation)cascaded LUVE · 2K ablationFID_patch: 46.48 -> 47.03Hurt
add low-frequency expert (ablation)LUVE 2K · 2K ablationFID_patch / Realism: FID_patch 43.86 -> 41.03; Realism 7.08 -> 7.64Improved
add high-frequency expert (ablation)LUVE 2K · 2K ablationFID_patch / AQ: FID_patch 44.44 -> 41.03; AQ 59.34 -> 59.78Improved
add HPS v3 data selection (ablation)LUVE 2K · 2K ablationFID_patch: 43.77 -> 41.03Improved
add Unsharp Masking augmentation for HFE data (ablation)LUVE 2K · 2K ablationFID_patch: 42.96 -> 41.03Improved
add decoder in VLUer (ablation)VLUer · qualitativeblurrinessImproved
add pixel-space and frame-difference loss to latent L1 loss (ablation)VLUer · qualitativeblocky artifactsImproved
replace training-free/VSR approaches with LUVEWan2.1-1.3B · VBenchaverage score: 83.75 (UltraWan-4K) -> 84.34 (Ours-2K)Improved

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy