AI Paper Atlas

Research map / Image generation / Super-resolution

Improved Adversarial Diffusion Compression for Real-World Video Super-Resolution

Bin Chen, Weiqi Li, Shijie Zhao, Xuanyu Zhang, Junlin Li, Li zhang, Jian Zhang

ICLR 2026 · Image generation › Super-resolution · approach: Diffusion-based super-resolution · 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
add 1D temporal convolutions to the 2D backbone (versus 3D DiT teacher size) (ablation)AdcVSR · UDM10/YouHQ40 and efficiencyparameters and E_warp: narrows DISTS gap to 3D model to 0.0014 with 7% parametersImproved
add temporal modeling (1D convs) to the 2D SD backbone AdcSR (ablation)AdcVSR · Real-VSR benchmarksvideo quality and E_warpImproved
double the number of 1D convolutions (ablation)AdcVSR · Real-VSR benchmarksvideo qualityNo clear effect
replace 1D convolutions with temporal self-attention (ablation)AdcVSR · Real-VSR benchmarksquality and lossNo clear effect
use dual heads instead of a single head (with dual-domain) (ablation)AdcVSR discriminator · YouHQ40E_warp: 6.32 -> 2.22Improved
add pixel-domain discriminator to dual-head feature-domain (ablation)AdcVSR discriminator · YouHQ40CLIPIQA: 0.6421 -> 0.6861Improved
add adversarial loss to distillation (ablation)AdcVSR · MVSR4xLPIPS: 0.3596 -> 0.3337Improved
add DOVE teacher guidance versus GT only (ablation)AdcVSR · MVSR4xMUSIQ: 50.32 -> 61.48Improved
use DOVE as teacher instead of SeedVR2 (ablation)AdcVSR · MVSR4xLPIPS: 0.3489 -> 0.3337Improved
use DOVE as teacher instead of DLoRAL (ablation)AdcVSR · MVSR4xMUSIQ: 54.61 -> 61.48Improved
share backbone and tail convolutions between detail and consistency heads (ablation)AdcVSR discriminator · training stabilitytraining convergenceImproved
add shuffled-video and assembled-image negatives to head training data (ablation)AdcVSR dual-head · Real-VSRE_warp and LPIPSImproved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 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