Research map / Video / Video generation
ProPhy: Progressive Physical Alignment for Dynamic World Simulation
Zijun Wang, Panwen Hu, Jing Wang, Terry Jingchen Zhang, Yuhao Cheng, Long Chen, Yiqiang Yan, Zutao Jiang, Hanhui Li, Xiaodan Liang
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 changed | Where | Measured | Effect |
|---|---|---|---|
| add ProPhy | Wan2.1-1.3B · VideoPhy2 (ALL) | Joint: 24.8 -> 26.5 | Improved |
| add ProPhy | CogVideoX-5B · VideoPhy2 (ALL) | Joint: 22.3 -> 26.7 | Improved |
| add ProPhy instead of WISA | CogVideoX-5B · VideoPhy2 (ALL) | Joint: 25.8 -> 26.7 | Improved |
| add ProPhy instead of VideoREPA | CogVideoX-5B · VideoPhy2 (ALL) | SA: 24.2 -> 32.2 | Improved |
| add ProPhy | CogVideoX-5B · VBench (600 prompts) | Dynamic Degree: 46.8 -> 72.0 | Improved |
| add ProPhy | CogVideoX-5B · VBench (600 prompts) | Aesthetic Quality: 48.2 -> 47.6 | Hurt |
| add ProPhy | Wan2.1-1.3B · VBench (600 prompts) | Quality Score: 77.3 -> 79.0 | Improved |
| use Physical Branch instead of LoRA on the backbone (ablation) | SEB+REB on Wan2.1-1.3B · VideoPhy2 (ALL, ablation) | Joint: 26.0 -> 26.5 | Improved |
| add REB (ablation) | Physical Blocks (no SEB) · VideoPhy2 (ALL, ablation) | Joint: 25.7 -> 26.2 | Improved |
| add SEB (ablation) | Physical Blocks (no REB) · VideoPhy2 (ALL, ablation) | Joint: 25.7 -> 26.0 | Improved |
| add SEB (ablation) | Physical Blocks + REB · VideoPhy2 (ALL, ablation) | Joint: 26.2 -> 26.5 | Improved |
| use relative-distance loss in SEB instead of BCE (ablation) | ProPhy on Wan2.1-1.3B · VideoPhy2 (ALL, ablation) | Joint: 26.3 -> 26.5 | Improved |
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.
- CogVideoX-5B — CogVideoX: Text-to-Video Diffusion Models with An Expert Transformer (ICLR 2025) · better on 4 of 4 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
- Physics-Informed Video Generation via Mixture-of-Experts Latent Alignment (NeurIPS 2026 · similarity 0.91)
- WISA: World simulator assistant for physics-aware text-to-video generation (NeurIPS 2025 · similarity 0.89)
- VideoREPA: Learning Physics for Video Generation through Relational Alignment with Foundation Models (NeurIPS 2025 · similarity 0.88)
- Hierarchical Fine-grained Preference Optimization for Physically Plausible Video Generation (NeurIPS 2025 · similarity 0.87)
- PhysCtrl: Generative Physics for Controllable and Physics-Grounded Video Generation (NeurIPS 2025 · similarity 0.86)
- PhyT2V: LLM-Guided Iterative Self-Refinement for Physics-Grounded Text-to-Video Generation (CVPR 2025 · similarity 0.86)
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.
