AI Paper Atlas

Research map / Video / Video generation

VPO: Aligning Text-to-Video Generation Models with Prompt Optimization

Jiale Cheng, Ruiliang Lyu, Xiaotao Gu, Xiao Liu, Jiazheng Xu, Yida Lu, Jiayan Teng, Zhuoyi Yang, Yuxiao Dong, Jie Tang, Hongning Wang, Minlie Huang

ICCV 2025 · Video › Video generation · approach: Compositional text-to-video diffusion · 21 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 preference optimization (DPO) after SFT (ablation)VPO-SFT · MonetBench, CogVideoX-2BOverall: 3.59 -> 3.76Improved
add preference optimization (DPO) after SFT (ablation)VPO-SFT · MonetBench, CogVideoX-5BOverall: 4.01 -> 4.15Improved
add text-level feedback to video-level feedback (ablation)VPO w/o TL FDBK · MonetBench, CogVideoX-2BOverall: 3.72 -> 3.76Improved
add text-level feedback to video-level feedback (ablation)VPO w/o TL FDBK · MonetBench, CogVideoX-5BOverall: 4.12 -> 4.15Improved
add text-level feedback to video-level feedback (ablation)VPO w/o TL FDBK · Query alignment, CogVideoX-2BAligned (%): 93.0 -> 94.6Improved
add text-level feedback to video-level feedback (ablation)VPO w/o TL FDBK · Query alignment, CogVideoX-5BAligned (%): 92.8 -> 94.8Improved
add video-level-only preference optimization after SFT (ablation)VPO-SFT · Query alignment safety (unsafe rate, CogVideoX-2B)Unsafe (%): 0.8 -> 2.0Hurt
use VPO instead of GPT-4o few-shot prompt rewritingGPT-4o Few-Shot · Query alignmentAligned (%): 86.4 -> 94.6Improved
use Promptist image prompt optimizer instead of original queryOriginal Query · MonetBench, CogVideoX-2BOverall: 3.27 -> 2.87Hurt
use VPO instead of GPT-4o few-shot rewritingGPT-4o Few-Shot · MonetBench, CogVideoX-5BOverall: 4.03 -> 4.15Improved
apply prompt optimizer trained on CogVideoX-2B to another video modelGPT-4o Few-Shot · MonetBench, Open-Sora 1.2Overall: 3.07 -> 3.18Improved
use VPO instead of VBench Long PromptsVBench Long Prompts (2B) · VBench Multiple Objectsscore (%): 63.81 -> 70.17Improved

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