AI Paper Atlas

Research map / Video / Video generation

InfVSR: Toward Consistency-Driven Streaming Generative Video Super-Resolution

Ziqing Zhang, Kai Liu, Zheng Chen, Xi Li, Yucong Chen, Bingnan Duan, Linghe Kong, Yulun Zhang

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
use autoregressive inference with rolling KV-cache instead of simple chunking (ablation)InfVSR · UDM10LPIPS: 0.3178 -> 0.2972Improved
add overlap and blending aggregation to chunking instead of AR (ablation)chunked VSR · UDM10E*warp / semantic consistency: E*warp 2.20 -> 1.96; (BC+SC)/2 0.9456 -> 0.9456Mixed
add joint visual guidance (vs none) (ablation)InfVSR · UDM10DISTS: 0.1518 -> 0.1422Improved
share one guidance across chunks (joint) instead of separate per chunk (ablation)InfVSR · UDM10(BC+SC)/2: 0.9547 -> 0.9578Improved
use chunk/cache (3,3) instead of (1,1) (ablation)InfVSR · UDM10CLIP-IQA: 0.4755 -> 0.5142Improved
increase chunk/cache from (3,3) to (5,5) (ablation)InfVSR · UDM10PSNR: 24.86 -> 24.90No clear effect
keep full KV-cache (infinity,3) instead of rolling length 3 (ablation)InfVSR · UDM10CLIP-IQA: 0.5142 -> 0.5084Hurt
add temporal smoothness loss Ltemp (ablation)InfVSR · UDM10E*warp: 2.23 -> 1.95Improved
add patch-wise pixel supervision (ablation)InfVSR · UDM10CLIP-IQA: 0.3997 -> 0.5142Improved
add Stage-I high-resolution pretraining (ablation)InfVSR · UDM10LPIPS: 0.3125 -> 0.2972Improved
add cross-chunk DMD loss (ablation)InfVSR · UDM10CLIP-IQA / DOVER: CLIP-IQA 0.5028 -> 0.5142; DOVER 0.7603 -> 0.7826; PSNR 25.04 -> 24.86Mixed
replace Ltemp with optical-flow alignment loss (ablation)InfVSR · UDM10E*warp: 1.95 -> 2.24Hurt

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