# ROWBench: Do Video Models Render What the Program Specifies?

arXiv: https://arxiv.org/abs/2610.02205 · HTML: https://mlsanity.com/paper/2610.02205/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Placed in Video-generation world models and simulators, one of 11 directions in World models.
- Builds on 23 of its 63 references that are in the map.
- First posted on arXiv 1 Oct 2026.

## Placed in

- World models / [Video-generation world models and simulators](https://mlsanity.com/wm/video-generation-world-models-and-simulators/)

## Builds on (most cited first, 23 of 23)

- [World Simulation with Video Foundation Models for Physical AI](https://mlsanity.com/paper/2511.00062/) (2025)
- [ViPE: Video Pose Engine for 3D Geometric Perception](https://mlsanity.com/paper/2508.10934/) (2025)
- [WorldPlay: Towards Long-Term Geometric Consistency for Real-Time Interactive World Modeling](https://mlsanity.com/paper/2512.14614/) (2025)
- [Advancing Open-source World Models](https://mlsanity.com/paper/2601.20540/) (2026)
- [CamI2V: Camera-Controlled Image-to-Video Diffusion Model](https://mlsanity.com/paper/2410.15957/) (2024)
- [Matrix-Game 3.0: Real-Time and Streaming Interactive World Model with Long-Horizon Memory](https://mlsanity.com/paper/2604.08995/) (2026)
- [SANA-WM: Efficient Minute-Scale World Modeling with Hybrid Linear Diffusion Transformer](https://mlsanity.com/paper/2605.15178/) (2026)
- [WBench: A Comprehensive Multi-turn Benchmark for Interactive Video World Model Evaluation](https://mlsanity.com/paper/2605.25874/) (2026)
- [MIND: Benchmarking Memory Consistency and Action Control in World Models](https://mlsanity.com/paper/2602.08025/) (2026)
- [WorldMark: A Unified Benchmark Suite for Interactive Video World Models](https://mlsanity.com/paper/2604.21686/) (2026)
- [Matrix-Game: Interactive World Foundation Model](https://mlsanity.com/paper/2506.18701/) (2025)
- [Omni-WorldBench: Towards a Comprehensive Interaction-Centric Evaluation for World Models](https://mlsanity.com/paper/2603.22212/) (2026)
- [Code World Model: Coding Agent as World Brain](https://mlsanity.com/paper/2608.25927/) (2026)
- [EchoWM: Open and Enterable Omnimodal World Models](https://mlsanity.com/paper/2608.23189/) (2026)
- [MemoBench: Benchmarking World Modeling in Dynamically Changing Environments](https://mlsanity.com/paper/2606.27537/) (2026)
- [AlayaWorld: Interactive Long-Horizon World Modeling -- Full Technical Report](https://mlsanity.com/paper/2607.18367/) (2026)
- [Generative World Renderer](https://mlsanity.com/paper/2604.02329/) (2026)
- [Programmable World Model](https://mlsanity.com/paper/2609.10540/) (2026)
- [Coarse-to-Real: Generative Rendering for Populated Dynamic Scenes](https://mlsanity.com/paper/2601.22301/) (2026)
- [PlayWorld: Benchmarking World Models with Agent Players over Long-Horizon Objectives](https://mlsanity.com/paper/2608.13552/) (2026)
- [Generative World Renderer at the Speed of Play](https://mlsanity.com/paper/2607.18703/) (2026)
- [LynnReal-Omni: Native multi-modal Video Generation for Agentic Visual Workflows](https://mlsanity.com/paper/2609.15863/) (2026)
- [TriWorldBench: A Tri-View Consistency Perspective on Embodied World Models](https://mlsanity.com/paper/2609.26314/) (2026)

## Abstract (arXiv)

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.