About

A physics-grounded benchmark for video generation. Browse model outputs by physical law, compare side-by-side, and explore evaluator-by-dataset leaderboards.

Citation

@misc{lin2026phygroundbenchmarkingphysicalreasoning,
  title         = {PhyGround: Benchmarking Physical Reasoning in Generative World Models},
  author        = {Juyi Lin and Arash Akbari and Yumei He and Lin Zhao and Haichao Zhang and Arman Akbari and Xingchen Xu and Zoe Y. Lu and Enfu Nan and Hokin Deng and Edmund Yeh and Sarah Ostadabbas and Yun Fu and Jennifer Dy and Pu Zhao and Yanzhi Wang},
  year          = {2026},
  eprint        = {2605.10806},
  archivePrefix = {arXiv},
  primaryClass  = {cs.CV},
  url           = {https://arxiv.org/abs/2605.10806}
}

Resources

• Paper (arXiv)
• GitHub repository
• Dataset — videos and first frames on HuggingFace
• Model — PhyJudge-9B evaluator weights on HuggingFace

License & contact

All content (code, generated videos, hand-curated annotations): CC-BY-4.0.

Questions / corrections: open an issue on the GitHub repo, or contact the corresponding author listed in the paper.

Build provenance

Built 2026-05-10T19:54:00Z