# Long-Horizon Consistent and Interaction-Aware World Models for Multi-Style End-to-End Driving

arXiv: https://arxiv.org/abs/2609.03225 · HTML: https://mlsanity.com/paper/2609.03225/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Placed in World models for autonomous driving, one of 11 directions in World models.
- Builds on 6 of its 40 references that are in the map.
- First posted on arXiv 3 Sep 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/)
- Deep reinforcement learning methods / [RL for autonomous driving and traffic](https://mlsanity.com/deeprl/rl-for-autonomous-driving-and-traffic/)

## Builds on (most cited first, 6 of 6)

- [Proximal Policy Optimization Algorithms](https://mlsanity.com/paper/1707.06347/) (2017)
- [DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models](https://mlsanity.com/paper/2402.03300/) (2024)
- [Lyapunov-based Safe Policy Optimization for Continuous Control](https://mlsanity.com/paper/1901.10031/) (2019)
- [EponaV2: Driving World Model with Comprehensive Future Reasoning](https://mlsanity.com/paper/2605.14696/) (2026)
- [Poutine: Vision-Language-Trajectory Pre-Training and Reinforcement Learning Post-Training Enable Robust End-to-End Autonomous Driving](https://mlsanity.com/paper/2506.11234/) (2025)
- [GraphWorld: Long-Horizon Planning with World Models for End-to-End Autonomous Driving](https://mlsanity.com/paper/2606.16274/) (2026)

## Abstract (arXiv)

End-to-end autonomous driving has increasingly adopted world model-based reinforcement learning frameworks to improve learning efficiency through textitimagined rollouts. However, existing world models suffer from three key limitations: temporal inconsistency in long-horizon imagined rollouts, inadequate modeling of ego-environment interactions, and limited adaptability to diverse driving styles. To address these challenges, we propose textitStyleDrive, a world-model-based learning framework that jointly enforces long-horizon consistency, explicitly disentangles interactive traffic states, and supports multi-style policy optimization within a unified learning paradigm. First, we introduce a temporal consistency regularization that integrates historical latent states through gated cross-attention, stabilizing long-horizon imagined rollouts and mitigating error accumulation. Second, we design an explicit state disentanglement module that separates ego-relevant from ego-irrelevant interactive states, enabling more interpretable and efficient decision-making in complex traffic scenarios. Third, we enable multi-style driving behaviors through Group Relative Policy Optimization, which replaces per-step reward optimization with trajectory-wise relative advantages, reducing reward variance and supporting diverse driving styles without retraining. We evaluate StyleDrive on the Bench2Drive closed-loop driving benchmark, achieving a driving score of 88.44 (+17.08 over the previous best world model-based method) and a success rate of 66.82 (+16.58). Furthermore, we deploy StyleDrive on a real automated guided vehicle platform and demonstrate promising sim-to-real transfer capability in dynamic driving scenarios.