PhysEditWorld - A Large-Scale Dataset Toward Physics-Editable World Models: Monitoring Performance and Model Drift

A result that is credible at launch may degrade as inputs, workflows, and populations change, making longitudinal monitoring part of the evidence rather than an afterthought. This structured evidence review evaluates "PhysEditWorld: A Large-Scale Dataset Toward Physics-Editable World Models" alongside nine author-disjoint, topically matched publications in physics-aware world modeling. It compares construct definitions, evaluation choices, operating assumptions, and reported limitations instead of treating bibliographic similarity as empirical equivalence. Viewed through longitudinal monitoring and model drift, the map separates claims supported by the available record from questions that still require full-text extraction, replication, or new experiments. The synthesis is interpretive rather than meta-analytic and therefore does not present a pooled effect estimate or a new causal result. The resulting agenda defines drift indicators, review intervals, alert thresholds, and criteria for recalibration, retraining, or retirement.

1. Hu, B., Ma, Y., Huang, J., Zhang, Z., Wu, H., Zhang, R., Li, Y., Wang, Z., Zhang, Y., Tseng, C.-M., Li, H., Qian, S., Zhou, J., Zhang, K., Liang, X., Jia, J., & Li, X. (2026). PhysEditWorld: A Large-Scale Dataset Toward Physics-Editable World Models. arXiv. https://doi.org/10.48550/arXiv.2606.26694 DOI
2. He, Y., Yuan, Z., Tu, Z., Ye, Y., & Sun, L. (2026). 3D4D: An Interactive, Editable, 4D World Model via 3D Video Generation. Proceedings of the AAAI Conference on Artificial Intelligence, 40(48), 41595-41597. https://doi.org/10.1609/aaai.v40i48.42351 DOI
3. Li, J. (2025). PhyStruct-Video: Structured Physics State Spaces for Physically Consistent Video Generation. . https://doi.org/10.36227/techrxiv.176538483.33325786/v1 DOI
4. Károly, A.-I., Nádas, I., & Galambos, P. (2024). Synthetic Multimodal Video Benchmark (SMVB): Utilizing Blender for rich dataset generation. 2024 IEEE 22nd World Symposium on Applied Machine Intelligence and Informatics (SAMI), 000065-000070. https://doi.org/10.1109/sami60510.2024.10432848 DOI
5. Namitha, K., Narayanan, A., & Geetha, M. (2020). A Synthetic Video Dataset Generation Toolbox for Surveillance Video Synopsis Applications. 2020 International Conference on Communication and Signal Processing (ICCSP), 493-497. https://doi.org/10.1109/iccsp48568.2020.9182084 DOI
6. Li, Q., Wu, R., & Xu, W. (2026). IMU2Video: World Model-Enabled Video Generation from IMU sensors. Proceedings of the 24th Annual International Conference on Mobile Systems, Applications and Services Workshops, 160-165. https://doi.org/10.1145/3812836.3814749 DOI
7. Li, K., Zhang, S., Fang, Y., Yuan, S., Zou, Y., & Yang, L. (2026). Towards Dynamic World Model Generation with Monocular Video. ICASSP 2026 - 2026 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), 12012-12016. https://doi.org/10.1109/icassp55912.2026.11463651 DOI
8. Chen, J., Wee, L., Dekker, A., & Bermejo, I. (2023). Using 3D deep features from CT scans for cancer prognosis based on a video classification model: A multi‐dataset feasibility study. Medical Physics, 50(7), 4220-4233. https://doi.org/10.1002/mp.16430 DOI
9. Wang, X., Wu, Z., & Peng, P. (2026). Fine-flow Distilling Coarse-flow Video Generation for Long-Term Driving World Model. Proceedings of the AAAI Conference on Artificial Intelligence, 40(31), 26526-26534. https://doi.org/10.1609/aaai.v40i31.39860 DOI
10. Zong, X., Tian, Y., Zhao, H., Zhu, S., Zhao, Z., & Zhang, H. (2025). LiveForgery: A Multi-Model Generation Synthetic Dataset for Deepfake Video Detection in Face Verification Scenario. 2025 9th IEEE International Conference on Network Intelligence and Digital Content (IC-NIDC), 363-367. https://doi.org/10.1109/ic-nidc67200.2025.11390370 DOI