AI Paper Atlas

Research map / Video / Video generation

Error Analyses of Auto-Regressive Video Diffusion Models

Jing Wang, Fengzhuo Zhang, Xiaoli Li, Vincent Y.~ F. Tan, Tianyu Pang, Chao Du, Aixin Sun, Zhuoran Yang

JMLR 2026 · Video › Video generation · approach: Efficient video diffusion models

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 history length from 16 to 112 frames (ablation)Prepending AR-VDM · DMLab history retrievalSRR: 0.18 -> 0.33Improved
increase history length from 16 to 112 frames (ablation)Prepending AR-VDM · Minecraft history retrievalSSIM: 0.50 -> 0.62Improved
increase compression budget from 8 to 48 history latents (ablation)Prepending AR-VDM · DMLab history retrievalSRR: 0.17 -> 0.25Improved
increase compression budget from 8 to 48 history latents (ablation)Prepending AR-VDM · Minecraft history retrievalSSIM: 0.51 -> 0.56Improved
increase history length from 16 to 112 frames (ablation)Channel Concat AR-VDM · DMLab history retrievalSRR: 0.09 -> 0.11Improved
use prepending instead of channel concatenationAR-VDM history conditioning · DMLab history retrieval (history 48)SRR: 0.12 -> 0.25Improved
add soft memory compression instead of last 16 framesPrepending AR-VDM · DMLab history retrieval, overallSRR: 0.260 -> 0.305Improved
add soft memory compression instead of FIFOPrepending AR-VDM · Minecraft history retrieval, overallSSIM: 0.405 -> 0.552Improved
add soft memory compression instead of last 32 frames with skipPrepending AR-VDM · Minecraft history retrieval, overallSSIM: 0.472 -> 0.552Improved
increase history length from 16 to 112 (ablation)Prepending AR-VDM · DMLab FVDFVD: 302.59 -> 308.81No clear effect
increase compression budget from 8 to 48 (ablation)Prepending AR-VDM · DMLab frames 48-64PSNR: 7.88 -> 6.76Hurt
increase compression budget from 8 to 48 (ablation)Prepending AR-VDM · Minecraft frames 48-64PSNR: 10.29 -> 9.47Hurt

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