World Embedding Benchmark
Physical fidelity has received increasing attention in world models and video generation, yet how video representations encode physical information remains less understood. We introduce the World Embedding Benchmark, comprising 8,000 controlled simulation cases from 80 families spanning fluid mechanics, solid mechanics, dynamics, and optics & electromagnetism. Each case pairs a rendered video with simulation-derived physical annotations, supporting three complementary tasks: text-video retrieval, physical-property regression, and multiple-choice video-description pair classification. We use these tasks to distinguish cross-modal physical alignment from the recoverability of quantitative p...
| Authors | Yiqi Liu, Ruifeng Yuan, Yang Wang et al. |
|---|---|
| Primary category | cs.CV |
| Published | 2 October 2026 |
| Last updated at source | 2 October 2026 |
Models and datasets referencing this paper
- World-Representation-Lab/World-Embedding-Optics-RetrievalDataset
- World-Representation-Lab/World-Embedding-Dynamics-RetrievalDataset
- World-Representation-Lab/World-Embedding-Solid-RetrievalDataset
- World-Representation-Lab/World-Embedding-RegressionDataset
- World-Representation-Lab/World-Embedding-Fluid-RetrievalDataset
Silytics shows public metadata from arXiv and links to the original. Information may be incomplete or out of date. Last refreshed 16 October 2026. Licenses and terms of the original apply to the work itself. Not financial, legal or professional advice.