AI Paper Atlas

Research map / Video / Video generation

Ref4D-VideoBench: Four-Dimensional Reference-Based Evaluation of Text-to-Video Generative Models

Jiajia Wei, Yujia He, Yuhan Hou, Hang Qi, Sihua Wang, Jincheng Shi, Kwok Fung Li, Zibin Zheng, Weibin Wu

CVPR 2026 · Video › Video generation · approach: Compositional text-to-video diffusion

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 per-video rule/question bank (ablation)MiniCPM-V-4.5 world-knowledge scorer · World knowledge consistency vs humanSRCC: 0.413 -> 0.847Improved
add per-video rule/question bank (ablation)VideoLLaMA3-7B world-knowledge scorer · World knowledge consistency vs humanSRCC: 0.406 -> 0.757Improved
use MiniCPM-V-4.5 instead of VideoLLaMA3-7B as judge (ablation)world-knowledge scorer · World knowledge consistency vs humanSRCC: 0.757 -> 0.847Improved
use reference-based framework instead of best no-reference baselineRef4D · Human alignment: SemanticSRCC: 0.317 -> 0.822Improved
use reference-based framework instead of best baselineRef4D · Human alignment: MotionSRCC: 0.358 -> 0.659Improved
use reference-based framework instead of best baselineRef4D · Human alignment: EventSRCC: 0.220 -> 0.755Improved
use reference-based framework instead of best baselineRef4D · Human alignment: World knowledgeSRCC: 0.391 -> 0.847Improved
use learned linear aggregator over atomic metrics (semantic) (ablation)Ref4D · Human alignment: SemanticSRCC: 0.734 -> 0.822Improved
use learned linear aggregator over atomic metrics (event) (ablation)Ref4D · Human alignment: EventSRCC: 0.718 -> 0.755Improved
use learned linear aggregator over atomic metrics (motion) (ablation)Ref4D · Human alignment: MotionSRCC: -0.498 (Dmag) -> 0.659Improved

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