# Beyond Pixel Histories: World Models with Persistent 3D State

arXiv: https://arxiv.org/abs/2603.03482 · HTML: https://mlsanity.com/paper/2603.03482/

## Key facts

- Cited by 13 papers in the map (data through 14 Oct 2026).
- Placed in Video-generation world models and simulators, one of 11 directions in World models.
- Builds on 41 of its 55 references that are in the map.
- First posted on arXiv 3 Mar 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, 30 of 41)

- [An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale](https://mlsanity.com/paper/2010.11929/) (2020)
- [Scalable Diffusion Models with Transformers](https://mlsanity.com/paper/2212.09748/) (2022)
- [Flow Matching for Generative Modeling](https://mlsanity.com/paper/2210.02747/) (2022)
- [RoFormer: Enhanced Transformer with Rotary Position Embedding](https://mlsanity.com/paper/2104.09864/) (2021)
- [Instant Neural Graphics Primitives with a Multiresolution Hash Encoding](https://mlsanity.com/paper/2201.05989/) (2022)
- [Quo Vadis, Action Recognition? A New Model and the Kinetics Dataset](https://mlsanity.com/paper/1705.07750/) (2017)
- [VGGT: Visual Geometry Grounded Transformer](https://mlsanity.com/paper/2503.11651/) (2025)
- [World Models](https://mlsanity.com/paper/1803.10122/) (2018)
- [Matterport3D: Learning from RGB-D Data in Indoor Environments](https://mlsanity.com/paper/1709.06158/) (2017)
- [Towards Accurate Generative Models of Video: A New Metric & Challenges](https://mlsanity.com/paper/1812.01717/) (2018)
- [On the Continuity of Rotation Representations in Neural Networks](https://mlsanity.com/paper/1812.07035/) (2018)
- [Building Normalizing Flows with Stochastic Interpolants](https://mlsanity.com/paper/2209.15571/) (2022)
- [Structured 3D Latents for Scalable and Versatile 3D Generation](https://mlsanity.com/paper/2412.01506/) (2024)
- [ScanNet++: A High-Fidelity Dataset of 3D Indoor Scenes](https://mlsanity.com/paper/2308.11417/) (2023)
- [Diffusion Forcing: Next-token Prediction Meets Full-Sequence Diffusion](https://mlsanity.com/paper/2407.01392/) (2024)
- [Self Forcing: Bridging the Train-Test Gap in Autoregressive Video Diffusion](https://mlsanity.com/paper/2506.08009/) (2025)
- [MineDojo: Building Open-Ended Embodied Agents with Internet-Scale Knowledge](https://mlsanity.com/paper/2206.08853/) (2022)
- [Neuralangelo: High-Fidelity Neural Surface Reconstruction](https://mlsanity.com/paper/2306.03092/) (2023)
- [Video PreTraining (VPT): Learning to Act by Watching Unlabeled Online Videos](https://mlsanity.com/paper/2206.11795/) (2022)
- [Light Field Networks: Neural Scene Representations with Single-Evaluation Rendering](https://mlsanity.com/paper/2106.02634/) (2021)
- [Transformers are Sample-Efficient World Models](https://mlsanity.com/paper/2209.00588/) (2022)
- [MobileNeRF: Exploiting the Polygon Rasterization Pipeline for Efficient Neural Field Rendering on Mobile Architectures](https://mlsanity.com/paper/2208.00277/) (2022)
- [Diffusion for World Modeling: Visual Details Matter in Atari](https://mlsanity.com/paper/2405.12399/) (2024)
- [Diffusion Models Are Real-Time Game Engines](https://mlsanity.com/paper/2408.14837/) (2024)
- [SAM 3D: 3Dfy Anything in Images](https://mlsanity.com/paper/2511.16624/) (2025)
- [MineRL: A Large-Scale Dataset of Minecraft Demonstrations](https://mlsanity.com/paper/1907.13440/) (2019)
- [Matrix-game 2.0: An open-source real-time and streaming interactive world model](https://mlsanity.com/paper/2508.13009/) (2025)
- [History-Guided Video Diffusion](https://mlsanity.com/paper/2502.06764/) (2025)
- [WorldMem: Long-term Consistent World Simulation with Memory](https://mlsanity.com/paper/2504.12369/) (2025)
- [Training Agents Inside of Scalable World Models](https://mlsanity.com/paper/2509.24527/) (2025)

## Built on by (most cited first, 13 of 13)

- [Towards Interactive Video World Modeling: Frontiers, Challenges, Benchmarks, and Future Trends](https://mlsanity.com/paper/2606.01164/) (2026)
- [Feed-Forward 3D Scene Modeling: A Problem-Driven Perspective](https://mlsanity.com/paper/2604.14025/) (2026)
- [Echo-Memory: A Controlled Study of Memory in Action World Models](https://mlsanity.com/paper/2606.09803/) (2026)
- [Marionette: Predicting World States, Rendering Geometry, Painting Appearance](https://mlsanity.com/paper/2608.14530/) (2026)
- [MilliVid: Hierarchical Latents for Long-Range Consistency in Video Generation](https://mlsanity.com/paper/2606.09056/) (2026)
- [AI for Games in the Foundation Model Era](https://mlsanity.com/paper/2609.16679/) (2026)
- [WorldAgents: Can Foundation Image Models be Agents for 3D World Models?](https://mlsanity.com/paper/2603.19708/) (2026)
- [Closing the Loop: Training-Free Revisit Consistency for Autoregressive Generative Rendering](https://mlsanity.com/paper/2607.21848/) (2026)
- [Dream-Cubed: Controllable Generative Modeling in Minecraft by Training on Billions of Cubes](https://mlsanity.com/paper/2604.22847/) (2026)
- [Valerant: An Automatic Navigable Game Map Generator via Action-Conditioned World Model Exploration](https://mlsanity.com/paper/2609.09418/) (2026)
- [Code Plans, Diffusion Renders: Open-Ended Generative World Modeling](https://mlsanity.com/paper/2609.26458/) (2026)
- [Learning What to Recall: Adaptive Multi-Cue Episodic Memory for World Models](https://mlsanity.com/paper/2609.34677/) (2026)
- [Honeycomb: Constant-Size Scene Memory Representation for Video World Models](https://mlsanity.com/paper/2609.37690/) (2026)

## Abstract (arXiv)

Interactive world models continually generate video by responding to a user's actions, enabling open-ended generation capabilities. However, existing models typically lack a 3D representation of the environment, meaning 3D consistency must be implicitly learned from data, and spatial memory is restricted to limited temporal context windows. This results in an unrealistic user experience and presents significant obstacles to downstream tasks such as training agents. To address this, we present PERSIST, a new paradigm of world model which simulates the evolution of a latent 3D scene: environment, camera, and renderer. This allows us to synthesise new frames with persistent spatial memory and consistent geometry. Both quantitative metrics and a qualitative user study show substantial improvements in spatial memory, 3D consistency, and long-horizon stability over existing methods, enabling coherent, evolving 3D worlds. We further demonstrate novel capabilities, including synthesising diverse 3D environments from a single image, as well as enabling fine-grained, geometry-aware control over generated experiences by supporting environment editing and specification directly in 3D space. Project page: https://francelico.github.io/persist.github.io