AI Paper Atlas

Research map / Video / Video generation

PhyGDPO: Physics-Aware Groupwise Direct Preference Optimization for Physically Consistent Text-to-Video Generation

Yuanhao Cai, Kunpeng Li, Menglin Jia, Jialiang Wang, Junzhe Sun, Feng Liang, Weifeng Chen, Felix Juefei-Xu, Chu Wang, Ali Thabet, Xiaoliang Dai, Xuan JU and 2 more

ECCV 2026 · Video › Video generation · approach: Compositional text-to-video diffusion · 27 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
post-train with PhyGDPOWan2.1-14B · VideoPhy2overall: 0.1288 -> 0.1627Improved
post-train with PhyGDPOWan2.1-14B · VideoPhy2 Hardscore: 0.0111 -> 0.0500Improved
add LoRA-SR (ablation)PhyGDPO · VideoPhy2overall: 0.1288 -> 0.1458Improved
add groupwise Plackett-Luce model (ablation)PhyGDPO · VideoPhy2overall: 0.1458 -> 0.1559Improved
add physics-guided rewarding (ablation)PhyGDPO · VideoPhy2overall: 0.1559 -> 0.1627Improved
add CoT data filtering in PhyAugPipe (ablation)PhyGDPO · VideoPhy2overall: 0.1475 -> 0.1525Improved
add action clustering in PhyAugPipe (ablation)PhyGDPO · VideoPhy2overall: 0.1525 -> 0.1575Improved
add physics-rewarded sampling in PhyAugPipe (ablation)PhyGDPO · VideoPhy2overall: 0.1575 -> 0.1627Improved
PhyGDPO instead of VideoDPOWan2.1-1.3B · VideoPhy2overall: 0.1305 -> 0.1407Improved
use LoRA-SR instead of full-copy reference (ablation)PhyGDPO · VideoPhy2GPU memory: 48.7GB -> 25.3GBImproved
use LoRA-SR instead of full-copy reference (ablation)PhyGDPO · VideoPhy2overall: 0.1373 -> 0.1407Improved
DPO with LoRA-SR instead of LoRA-SFT (ablation)Wan2.1-1.3B · VideoPhy2overall: 0.1283 -> 0.1407Improved

Later papers that compared with this method

1 later accepted paper reports PhyGDPO as a baseline in their results tables; 1 of them report a better number on most of the results they share with it.

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