AI Paper Atlas

Research map / Video / Video generation

RefAlign: Representation Alignment for Reference-to-Video Generation

Lei Wang, YuXin Song, Ge Wu, Haocheng Feng, Hang Zhou, Jingdong Wang, Yaxing Wang, Jian Yang

ECCV 2026 · Video › Video generation · approach: Compositional text-to-video diffusion · 10 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 RA loss (ablation)RefAlign (Wan2.1 1.3B) · OpenS2V-EvalTotalScore: 49.93% -> 55.73%Improved
add RA loss (ablation)RefAlign (Wan2.1 1.3B) · OpenS2V-EvalNaturalScore: 40.46% -> 62.96%Improved
add RA loss (ablation)RefAlign (Wan2.1 1.3B) · OpenS2V-EvalFaceSim: 68.45% -> 53.15%Hurt
add negative loss L_neg to RA loss (ablation)RefAlign · OpenS2V-EvalTotalScore: 51.75% -> 55.73%Improved
align to DINOv3 via loss instead of feeding DINOv3 as extra input (ablation)RefAlign · OpenS2V-EvalTotalScore: 52.15% -> 55.73%Improved
add DINOv3 input features (config D) vs no RA loss (ablation)Wan2.1 baseline · OpenS2V-EvalTotalScore: 49.93% -> 52.15%Improved
use DINOv3-H+ instead of DINOv3-L as alignment encoder (ablation)RefAlign · OpenS2V-EvalTotalScore: 55.73% -> 55.30%No clear effect
use SigLIP2-So instead of DINOv3-L as alignment encoder (ablation)RefAlign · OpenS2V-EvalTotalScore: 55.73% -> 54.25%Hurt
use Qwen2.5-VL-7B instead of DINOv3-L as alignment encoder (ablation)RefAlign · OpenS2V-EvalTotalScore: 55.73% -> 53.12%Hurt
increase alignment depth from 9 to 30 (ablation)RefAlign · OpenS2V-EvalFaceSim: 54.28% -> 9.59%Hurt
increase alignment depth from 9 to 30 (ablation)RefAlign · OpenS2V-EvalNaturalScore: 66.62% -> 80.21%Improved
increase alignment depth from 9 to 30 (ablation)RefAlign · OpenS2V-EvalTotalScore: 55.40% -> 50.12%Hurt

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