# ReinDriveGen: Reinforcement Post-Training for Out-of-Distribution Driving Scene Generation

arXiv: https://arxiv.org/abs/2604.01129 · HTML: https://mlsanity.com/paper/2604.01129/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 30 of its 31 references that are in the map.
- First posted on arXiv 1 Apr 2026.

## Builds on (most cited first, 30 of 30)

- [3D Gaussian Splatting for Real-Time Radiance Field Rendering](https://mlsanity.com/paper/2308.04079/) (2023)
- [Flow Matching for Generative Modeling](https://mlsanity.com/paper/2210.02747/) (2022)
- [ShapeNet: An Information-Rich 3D Model Repository](https://mlsanity.com/paper/1512.03012/) (2015)
- [Scalability in Perception for Autonomous Driving: Waymo Open Dataset](https://mlsanity.com/paper/1912.04838/) (2019)
- [VBench: Comprehensive Benchmark Suite for Video Generative Models](https://mlsanity.com/paper/2311.17982/) (2023)
- [Deformable 3D Gaussians for High-Fidelity Monocular Dynamic Scene Reconstruction](https://mlsanity.com/paper/2309.13101/) (2023)
- [Flow-GRPO: Training Flow Matching Models via Online RL](https://mlsanity.com/paper/2505.05470/) (2025)
- [VACE: All-in-One Video Creation and Editing](https://mlsanity.com/paper/2503.07598/) (2025)
- [DrivingGaussian: Composite Gaussian Splatting for Surrounding Dynamic Autonomous Driving Scenes](https://mlsanity.com/paper/2312.07920/) (2023)
- [Vista: A Generalizable Driving World Model with High Fidelity and Versatile Controllability](https://mlsanity.com/paper/2405.17398/) (2024)
- [UniSim: A Neural Closed-Loop Sensor Simulator](https://mlsanity.com/paper/2308.01898/) (2023)
- [PoinTr: Diverse Point Cloud Completion with Geometry-Aware Transformers](https://mlsanity.com/paper/2108.08839/) (2021)
- [DiffusionNFT: Online Diffusion Reinforcement with Forward Process](https://mlsanity.com/paper/2509.16117/) (2025)
- [Street Gaussians: Modeling Dynamic Urban Scenes with Gaussian Splatting](https://mlsanity.com/paper/2401.01339/) (2024)
- [MARS: An Instance-aware, Modular and Realistic Simulator for Autonomous Driving](https://mlsanity.com/paper/2307.15058/) (2023)
- [SUDS: Scalable Urban Dynamic Scenes](https://mlsanity.com/paper/2303.14536/) (2023)
- [NeuRAD: Neural Rendering for Autonomous Driving](https://mlsanity.com/paper/2311.15260/) (2023)
- [Periodic Vibration Gaussian: Dynamic Urban Scene Reconstruction and Real-time Rendering](https://mlsanity.com/paper/2311.18561/) (2023)
- [DriveDreamer4D: World Models Are Effective Data Machines for 4D Driving Scene Representation](https://mlsanity.com/paper/2410.13571/) (2024)
- [StreetSurf: Extending Multi-view Implicit Surface Reconstruction to Street Views](https://mlsanity.com/paper/2306.04988/) (2023)
- [textitS3Gaussian: Self-Supervised Street Gaussians for Autonomous Driving](https://mlsanity.com/paper/2405.20323/) (2024)
- [MagicDrive-V2: High-Resolution Long Video Generation for Autonomous Driving with Adaptive Control](https://mlsanity.com/paper/2411.13807/) (2024)
- [Pref-GRPO: Pairwise Preference Reward-based GRPO for Stable Text-to-Image Reinforcement Learning](https://mlsanity.com/paper/2508.20751/) (2025)
- [ReconDreamer: Crafting World Models for Driving Scene Reconstruction via Online Restoration](https://mlsanity.com/paper/2411.19548/) (2024)
- [YOLO26: Key Architectural Enhancements and Performance Benchmarking for Real-Time Object Detection](https://mlsanity.com/paper/2509.25164/) (2025)
- [FreeVS: Generative View Synthesis on Free Driving Trajectory](https://mlsanity.com/paper/2410.18079/) (2024)
- [StreetCrafter: Street View Synthesis with Controllable Video Diffusion Models](https://mlsanity.com/paper/2412.13188/) (2024)
- [FreeSim: Toward Free-viewpoint Camera Simulation in Driving Scenes](https://mlsanity.com/paper/2412.03566/) (2024)
- [MaskGWM: A Generalizable Driving World Model with Video Mask Reconstruction](https://mlsanity.com/paper/2502.11663/) (2025)
- [GA-Drive: Geometry-Appearance Decoupled Modeling for Free-viewpoint Driving Scene Generation](https://mlsanity.com/paper/2602.20673/) (2026)

## Abstract (arXiv)

We present ReinDriveGen, a framework that enables full controllability over dynamic driving scenes, allowing users to freely edit actor trajectories to simulate safety-critical corner cases such as front-vehicle collisions, drifting cars, vehicles spinning out of control, pedestrians jaywalking, and cyclists cutting across lanes. Our approach constructs a dynamic 3D point cloud scene from multi-frame LiDAR data, introduces a vehicle completion module to reconstruct full 360deg geometry from partial observations, and renders the edited scene into 2D condition images that guide a video diffusion model to synthesize realistic driving videos. Since such edited scenarios inevitably fall outside the training distribution, we further propose an RL-based post-training strategy with a pairwise preference model and a pairwise reward mechanism, enabling robust quality improvement under out-of-distribution conditions without ground-truth supervision. Extensive experiments demonstrate that ReinDriveGen outperforms existing approaches on edited driving scenarios and achieves state-of-the-art results on novel ego viewpoint synthesis.