# Toward Physically Consistent Driving Video World Models under Challenging Trajectories

arXiv: https://arxiv.org/abs/2603.24506 · HTML: https://mlsanity.com/paper/2603.24506/

## Key facts

- Cited by 4 papers in the map (data through 14 Oct 2026).
- Placed in World models for autonomous driving, one of 11 directions in World models.
- Builds on 17 of its 47 references that are in the map.
- First posted on arXiv 25 Mar 2026.

## Placed in

- World models / [World models for autonomous driving](https://mlsanity.com/wm/world-models-for-autonomous-driving/)
- End-to-end autonomous driving / [Generative simulation for driving](https://mlsanity.com/driving/generative-simulation-for-driving/)

## Builds on (most cited first, 17 of 17)

- [Wan: Open and Advanced Large-Scale Video Generative Models](https://mlsanity.com/paper/2503.20314/) (2025)
- [GAIA-1: A Generative World Model for Autonomous Driving](https://mlsanity.com/paper/2309.17080/) (2023)
- [CameraCtrl: Enabling Camera Control for Text-to-Video Generation](https://mlsanity.com/paper/2404.02101/) (2024)
- [World Simulation with Video Foundation Models for Physical AI](https://mlsanity.com/paper/2511.00062/) (2025)
- [ReCogDrive: A Reinforced Cognitive Framework for End-to-End Autonomous Driving](https://mlsanity.com/paper/2506.08052/) (2025)
- [ViPE: Video Pose Engine for 3D Geometric Perception](https://mlsanity.com/paper/2508.10934/) (2025)
- [WorldModelBench: Judging Video Generation Models As World Models](https://mlsanity.com/paper/2502.20694/) (2025)
- [BEVControl: Accurately Controlling Street-view Elements with Multi-perspective Consistency via BEV Sketch Layout](https://mlsanity.com/paper/2308.01661/) (2023)
- [DriveLaW:Unifying Planning and Video Generation in a Latent Driving World](https://mlsanity.com/paper/2512.23421/) (2025)
- [ReSim: Reliable World Simulation for Autonomous Driving](https://mlsanity.com/paper/2506.09981/) (2025)
- [Genesis: Multimodal Driving Scene Generation with Spatio-Temporal and Cross-Modal Consistency](https://mlsanity.com/paper/2506.07497/) (2025)
- [Challenger: Affordable Adversarial Driving Video Generation](https://mlsanity.com/paper/2505.15880/) (2025)
- [Rethinking Driving World Model as Synthetic Data Generator for Perception Tasks](https://mlsanity.com/paper/2510.19195/) (2025)
- [WorldSplat: Gaussian-Centric Feed-Forward 4D Scene Generation for Autonomous Driving](https://mlsanity.com/paper/2509.23402/) (2025)
- [SafeMVDrive: Multi-view Safety-Critical Driving Video Synthesis in the Real World Domain](https://mlsanity.com/paper/2505.17727/) (2025)
- [VILTA: A VLM-in-the-Loop Adversary for Enhancing Driving Policy Robustness](https://mlsanity.com/paper/2601.12672/) (2026)
- [Ctrl-Crash: Controllable Diffusion for Realistic Car Crashes](https://mlsanity.com/paper/2506.00227/) (2025)

## Built on by (most cited first, 4 of 4)

- [GigaWorld-1: A Roadmap to Build World Models for Robot Policy Evaluation](https://mlsanity.com/paper/2607.02642/) (2026)
- [Towards Interactive Video World Modeling: Frontiers, Challenges, Benchmarks, and Future Trends](https://mlsanity.com/paper/2606.01164/) (2026)
- [CoDrive: Cross-Vehicle World-Consistent Video Generation with Precise Trajectory Control for Cooperative Driving](https://mlsanity.com/paper/2609.34749/) (2026)
- [TerrainForge: Physics-Grounded Road Geometry Editing for Counterfactual Autonomous Driving](https://mlsanity.com/paper/2610.02825/) (2026)

## Abstract (arXiv)

Video generation models have shown strong potential as world models for autonomous driving simulation. However, existing approaches are primarily trained on real-world driving datasets, which mostly contain natural and safe driving scenarios. As a result, current models often fail when conditioned on challenging or counterfactual trajectories-such as imperfect trajectories generated by simulators or planning systems-producing videos with severe physical inconsistencies and artifacts. To address this limitation, we propose PhyGenesis, a world model designed to generate driving videos with high visual fidelity and strong physical consistency. Our framework consists of two key components: (1) a physical condition generator that transforms potentially invalid trajectory inputs into physically plausible conditions, and (2) a physics-enhanced video generator that produces high-fidelity multi-view driving videos under these conditions. To effectively train these components, we construct a large-scale, physics-rich heterogeneous dataset. Specifically, in addition to real-world driving videos, we generate diverse challenging driving scenarios using the CARLA simulator, from which we derive supervision signals that guide the model to learn physically grounded dynamics under extreme conditions. This challenging-trajectory learning strategy enables trajectory correction and promotes physically consistent video generation. Extensive experiments demonstrate that PhyGenesis consistently outperforms state-of-the-art methods, especially on challenging trajectories. Our project page is available at: https://wm-research.github.io/PhyGenesis/.