Metraipolis › archive › dc_6327d7ea8cc0
PROWBench: Do Video Models Render What the Program Specifies? — arXiv:2610.02205
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2610.02205
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-06T19:15:52.791Z
- SHA-256 of the archived text
66ab60537e6360cc3599141b80f4dc9a4f1e713a7c8fa17ca1cb64f484fafa57- Stable URL of this page
https://metraipolis.com/documento/dc_6327d7ea8cc0
How to cite
PROWBench: Do Video Models Render What the Program Specifies? — arXiv:2610.02205 — arXiv (en), https://arxiv.org/abs/2610.02205 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_6327d7ea8cc0 (Metraipolis archive dc_6327d7ea8cc0, sha256 66ab60537e6360cc3599141b80f4dc9a4f1e713a7c8fa17ca1cb64f484fafa57)
BibTeX
@misc{dc6327d7ea8cc0,
title = {PROWBench: Do Video Models Render What the Program Specifies? — arXiv:2610.02205},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2610.02205},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_6327d7ea8cc0; sha256 66ab60537e6360cc3599141b80f4dc9a4f1e713a7c8fa17ca1cb64f484fafa57},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.02205. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Programmable world models separate executable dynamics from visual generation, offering a promising foundation for next-generation game engines. However, their visual adherence to explicit rules and interactions remains insufficiently evaluated. Existing benchmarks assess visual quality, controllability, and instruction or physical adherence, but rarely test fidelity to fine-grained, program-specified world events. We introduce PROWBench, comprising 170 programmatically constructed episodes and 600 proxy videos covering diverse scenes and interactions. PROWBench logs entity states and timestamped events, including those outside the camera's field of view, as replayable world records, from which it renders synchronized views and proxy representations. This enables generated videos to be checked against the observable consequences of program execution. An extensible framework constructs scenes, controls behaviors, and can render each camera view in different representations, such as coarse 3D, and bounding boxes. The benchmark covers first- and third-person perspectives, with synchronized multi-view observations available for a subset of episodes. Grounded in these records, PROWBench evaluates entity control, long-horizon memory, and, with two VLM-based metrics, Logic-Render Alignment and Interaction Success Rate, adherence to the prescribed timeline and the visual realization of timestamped engine-recorded events. Tema degli esploratori: statistica e scienza dei dati (retro-catalogo arXiv: annunci del 2026-10-02 in cs.CV). Categorie arXiv: Computer Vision and Pattern Recognition (cs.CV). Pagina del lavoro: https://arxiv.org/abs/2610.02205 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito). Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina. Letto il 2026-10-06T17:37:36.900Z dalla pagina https://arxiv.org/catchup/cs.CV/2026-10-02?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).