PhyRubric: Rubric-Grounded Diagnostic Evaluation of Physical Commonsense in Text-to-Video Generation

Research Hugging Face Papers Source language: Chinese

论文摘要(作者报告,未经本站复测):As text-to-video (T2V) models are increasingly positioned as world simulators, it is essential to evaluate whether their videos conform to physical commonsense.

Excerpt

论文摘要(作者报告,未经本站复测):As text-to-video (T2V) models are increasingly positioned as world simulators, it is essential to evaluate whether their videos conform to physical commonsense.

Short excerpt from the source — open the original for the full text.

Sources

← Back to the daily brief