AI Paper Atlas

Research map / Video / Video generation

Mind the Generative Details: Direct Localized Detail Preference Optimization for Video Diffusion Models

Zitong Huang, Kaidong Zhang, Yukang Ding, Chao Gao, Rui Ding, Ying Chen, Wangmeng Zuo

CVPR 2026 · Video › Video generation · approach: Efficient video diffusion models · 7 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 region-aware DPO loss on top of DPO+SFT (ablation)LocalDPO (CogVideoX-2B) · VBench promptsImaging quality: 0.6514 -> 0.7080Improved
add region-aware DPO loss on top of DPO+SFT (ablation)LocalDPO (CogVideoX-2B) · VBench promptsAesthetic quality: 0.6308 -> 0.6499Improved
add region-aware DPO loss on top of DPO+SFT (ablation)LocalDPO (CogVideoX-2B) · VBench promptsVideoAlign Overall: 7.8003 -> 7.8568Improved
add SFT loss to DPO loss (ablation)LocalDPO (CogVideoX-2B) · VBench promptsAesthetic quality: 0.6303 -> 0.6308No clear effect
region-aware corruption negatives instead of vanilla-DPO lose samples with real positives (ablation)LocalDPO (CogVideoX-2B) · VBench promptsImaging quality: 0.6577 -> 0.7080Improved
region-aware corruption negatives instead of vanilla-DPO lose samples with real positives (ablation)LocalDPO (CogVideoX-2B) · VBench promptsVideoAlign Overall: 7.7778 -> 7.8568Improved
real-video positives with generated negatives instead of vanilla win/lose pairs (ablation)DPO (CogVideoX-2B) · VBench promptsVideoAlign Overall: 7.7862 -> 7.7778No clear effect
LocalDPO vs vanilla DPOCogVideoX-5B · VBench promptsVideoAlign Overall: 9.5910 -> 10.2930Improved
LocalDPO vs vanilla DPOCogVideoX-5B · VBench promptsAesthetic quality: 0.5953 -> 0.6274Improved
LocalDPO vs SFTWan2.1-1.3B · VBench promptsVideoAlign Overall: 7.7283 -> 7.9588Improved
vanilla DPO vs baselineCogVideoX-5B · VBench promptsAesthetic quality: 0.6110 -> 0.5953Hurt
LocalDPO vs vanilla DPOWan2.1-1.3B · VideoJAM benchmarkVideoAlign Overall: 7.0548 -> 7.4849Improved

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