AI Paper Atlas

Research map / Video / Video generation

Towards One-step Causal Video Generation via Adversarial Self-Distillation

Yongqi Yang, Huayang Huang, Xu Peng, Xiaobin Hu, Donghao Luo, Jiangning Zhang, Chengjie Wang, Yu Wu

ICLR 2026 · Video › Video generation · approach: Autoregressive long video generation · 13 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 adversarial self-distillation (ASD) training (ablation)few-step student · VBench 1-stepTotal Score: 78.13 -> 80.65Improved
add adversarial self-distillation (ASD) training (ablation)few-step student · VBench 1-stepSemantic Score: 69.31 -> 76.15Improved
add adversarial self-distillation (ASD) training (ablation)few-step student · VBench 2-stepTotal Score: 82.61 -> 83.28Improved
add first-frame enhancement (FFE) at inference (ablation)few-step student · VBench 1-stepTotal Score: 78.13 -> 83.04Improved
add first-frame enhancement (FFE) at inference (ablation)few-step student · VBench 1-stepSemantic Score: 69.31 -> 79.95Improved
add first-frame enhancement (FFE) at inference (ablation)few-step student · VBench 2-stepTotal Score: 82.61 -> 83.80Improved
add FFE on top of ASD (ablation)few-step student · VBench 1-stepTotal Score: 80.65 -> 83.89Improved
add FFE on top of ASD (ablation)few-step student · VBench 2-stepTotal Score: 83.28 -> 84.32Improved
use ASD+FFE instead of retrained Self ForcingSelf Forcing · VBench 1-stepTotal Score: 80.62 -> 83.89Improved
use ASD+FFE instead of Self Forcing at 4 stepsSelf Forcing · VBench 4-stepTotal Score: 84.31 -> 84.38No clear effect
set ASD loss weight alpha to zero (ablation)ASD loss · VBench 1* and 2* stepTotal ScoreImproved
add ASD loss to DMD training (ablation)DMD · DMD loss during trainingDMD loss variance: 0.01963 -> 0.00258Improved

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