Large-scale pre-trained video generation models excel in content creation but are not reliable as physically accurate world simulators out of the box. This work studies the process of post-training these models for accurate world modeling through the lens of the simple, yet fundamental, physics task of modeling object freefall. We show state-of-the-art video generation models struggle with this basic task, despite their visually impressive outputs. To remedy this problem, we find that fine-tuning on a relatively small amount of simulated videos is effective in inducing the dropping behavior in the model, and we can further improve results through a novel reward modeling procedure we introduce. Our study also reveals key limitations of post-training in generalization and distribution modeling. Additionally, we release a benchmark for this task that may serve as a useful diagnostic tool for tracking physical accuracy in large-scale video generative model development.
View on arXiv@article{li2025_2503.09595,
title={ PISA Experiments: Exploring Physics Post-Training for Video Diffusion Models by Watching Stuff Drop },
author={ Chenyu Li and Oscar Michel and Xichen Pan and Sainan Liu and Mike Roberts and Saining Xie },
journal={arXiv preprint arXiv:2503.09595},
year={ 2025 }
}