| title | 4DCodeBench: Benchmarking Agents on Inverse Graphics of Dynamic Scenes |
|---|---|
| authors | Ruihong Shen, Žiga Kovačič, Peter Kulits, Xingrui Wang, Zizhang Li, Joshua B. Tenenbaum, Alan Yuille, Jieneng Chen, Jiajun Wu |
| date | 2026-10-02 |
| venue | arXiv (preprint, not peer reviewed) |
| id | arXiv:2610.03715 |
| affiliation | 4DCodeBench |
| upvotes | 21 (Hugging Face Daily Papers, 2026-11-01) |
| links | abs pdf hf code |
| abstract | We introduce 4DCodeBench, a benchmark for 4D inverse graphics through code generation, in which agents reconstruct dynamic scenes from video as executable graphics programs. To accomplish this, agents must translate visual observations into compact representations of scene structure and dynamics, by implementing abstractions such as physical simulations to reproduce complex behavior. To evaluate this capability, we curate a set of real-world videos and construct synthetic scenes spanning diverse physical phenomena, including deformation, fluid flow, and fracture. We perform extensive benchmarking of frontier models, finding that strong static reconstruction capabilities do not yet translate into reliable reconstruction of complex dynamics. 4DCodeBench provides a testbed for tracking progress toward agents that can interpret the dynamics of the world through code. Our benchmark is available at https://github.com/4DCodeBench/4DCodeBench |
cite
@misc{shen2026dcodebench,
title={4DCodeBench: Benchmarking Agents on Inverse Graphics of Dynamic Scenes},
author={Ruihong Shen and Žiga Kovačič and Peter Kulits and Xingrui Wang and Zizhang Li and Joshua B. Tenenbaum and Alan Yuille and Jieneng Chen and Jiajun Wu},
year={2026},
eprint={2610.03715},
archivePrefix={arXiv},
url={https://arxiv.org/abs/2610.03715}
}