AI Paper Atlas

Research map / Video / Video generation

QuantCache: Adaptive Importance-Guided Quantization with Hierarchical Latent and Layer Caching for Video Generation

Junyi Wu, Zhiteng Li, Zheng Hui, Yulun Zhang, Linghe Kong, Xiaokang Yang

ICCV 2025 · Video › Video generation · approach: Efficient video diffusion models · 11 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
add hierarchical latent caching (HLC) (ablation)QuantCache on Open-Sora · Open-Sora promptsspeedup: 1.00x -> 4.12xImproved
add hierarchical latent caching (HLC) (ablation)QuantCache on Open-Sora · Open-Sora promptsaesthetic quality: 63.09 -> 62.09Hurt
add adaptive importance-guided quantization (AIGQ) (ablation)QuantCache on Open-Sora · Open-Sora promptsspeedup: 4.12x -> 6.33xImproved
add adaptive importance-guided quantization (AIGQ) (ablation)QuantCache on Open-Sora · Open-Sora promptsimaging quality: 58.28 -> 55.68Hurt
add structural redundancy-aware pruning (SRAP) (ablation)QuantCache on Open-Sora · Open-Sora promptsspeedup: 6.33x -> 6.72xImproved
add structural redundancy-aware pruning (SRAP) (ablation)QuantCache on Open-Sora · Open-Sora promptsbackground consistency: 97.62 -> 96.19Hurt
QuantCache versus AdaCache-fastQuantCache · Open-Sora speedupspeedup: 2.24x -> 6.72xImproved
QuantCache 4/6 versus ViDiT-Q 4/8QuantCache · VBenchaesthetic quality: 58.23 -> 58.63Improved
QuantCache 4/6 versus ViDiT-Q 4/8QuantCache · VBenchimaging quality: 57.21 -> 55.94Hurt
lower weight/activation bit-width for Q-DiT (8/8 to 4/8)Q-DiT · VBenchaesthetic quality: 58.68 -> 31.14Hurt
QuantCache versus ViDiT-Q at 8/8QuantCache · VBenchaesthetic quality: 59.89 -> 58.57Hurt
QuantCache 4/6 versus ViDiT-Q 4/8QuantCache · DOVERVQA technical: 49.11 -> 49.14No clear effect

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