AI Paper Atlas

Research map / Video / Video generation

ReHyAt: Recurrent Hybrid Attention for Video Diffusion Transformers

Mohsen Ghafoorian, Amirhossein Habibian

CVPR 2026 · Video › Video generation · approach: Efficient video diffusion models · 3 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
increase softmax chunk size Tc from 1 to 2 (ablation)ReHyAt (25 blocks) · VBench 320x480Total: 80.97 -> 82.08Improved
increase chunk size Tc from 2 to 5 (ablation)ReHyAt (25 blocks) · VBench 320x480Total: 82.08 -> 82.48Improved
enable chunk overlap To=1 versus 0 (ablation)ReHyAt (25 blocks, Tc=3) · VBench 320x480Total: 81.56 -> 82.17Improved
increase overlap from To=1 to 3 (ablation)ReHyAt · VBench 320x480Total: 82.17 -> 82.19No clear effect
make hybrid attention causal for RNN reformulation (ablation)ReHyAt 15xTc=3,To=0 · VBench 320x480Total: 82.27 -> 82.35No clear effect
distill Wan2.1 into ReHyAt 15xTc=3,To=1 (ablation)Wan2.1 1.3B · VBenchSemantic: 75.12 -> 80.70Improved
distill Wan2.1 into ReHyAt 15xTc=3,To=1Wan2.1 1.3B · VBenchQuality: 85.10 -> 84.57Hurt
ReHyAt versus original Wan2.1 in human preferenceWan2.1 1.3B · 500 paired comparisons% preferred: 27.6 vs 29.0 (43.5 no preference)No clear effect
ReHyAt versus Wan2.1 on VBench-2.0Wan2.1 1.3B · VBench-2.0Total: 56.0 -> 56.1No clear effect
replace flash attention with ReHyAt Tc=3Wan2.1 DiT blocks · Snapdragon8-Gen4, 121 frameslatency ms: 4809 -> 302Improved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 0 of 1 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

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