mlsanityData to 7 Oct 2026

Generative simulation for driving

255 papers1 this weekMedium confidence in this grouping: mostly one topic, with blurry edgesMedium confidence in this grouping: mostly one topic, with blurry edges · About

Keywords: driving, autonomous, autonomous driving, generation, video

Key facts. Generative simulation for driving: 255 papers on arXiv since 2021 (data through 7 Oct 2026). 64 of its 255 mapped papers (25%) cite “GAIA-1: A Generative World Model for Autonomous Driving” (2023, arXiv 2309.17080). 56 of its 255 mapped papers (22%) cite “Vista: A Generalizable Driving World Model with High Fidelity and Versatile Controllability” (2024, arXiv 2405.17398). 18 new papers 17 Sep–7 Oct.

What’s new

17 Sep – 7 Oct 2026

18 new papers, 17 Sep – 7 Oct; 1 this week.

  1. This week · 1
  2. Controllable and Photorealistic Pedestrian Risky Motion Generation for End-to-End Driving Safety Evaluation also fits: Adversarial scenarios and safety testing
  3. Earlier · 17
  4. MomWorld: Momentum-Aware Latent World Model for Long-Horizon Autonomous Driving
  5. PhysWAM: Physically Consistent World Action Model for Autonomous Driving
  6. World4Scorer: Outcome-Grounded World Modeling for Autonomous Driving
  7. CoDrive: Cross-Vehicle World-Consistent Video Generation with Precise Trajectory Control for Cooperative Driving
  8. AD-E2E-JEPA: A Joint-Embedding Predictive Architecture For End-to-End Autonomous Driving also fits: Imitation-learning end-to-end planners
  9. ReDrive: Shaping Representations with World Modeling for End-to-End Driving
  10. VehDyn: A Driving World Model Benchmark for Vehicle Dynamics
  11. OneFixer: High-Quality and Consistent One-Step Autoregressive 3DGS Refinement for Driving Scenes
  12. RECAST: From Log Replay to Closed-Loop Driving Simulation with View-Complete Actors
All 18
  1. WALT: Learning World-Model-Aligned Latent Trajectories for Autonomous Driving
  2. HelloWorld: Towards Practical Applications of Generative Driving World Models
  3. DreamStream: Towards Policy-Oriented Generative Simulation for End-to-End Driving
  4. ForeDrive: Foresight-Guided End-to-End Autonomous Driving with a Planning-Relevant Latent World Model
  5. Relationally Grounded Latent World Models for Autonomous Driving also fits: VLMs for autonomous driving
  6. DriveReferee: Geometric Safety Verdicts Need Not Be Learned for Driving World-Action Models
  7. ZYT-World: A Real-Time Controllable World Model for Closed-Loop Autonomous-Driving Simulation
  8. MM-Future: Multi-Mode Joint World-Action Modeling for Autonomous Driving

Foundations

Most cited by papers here

  1. GAIA-1: A Generative World Model for Autonomous Driving2023cited by 64 here
  2. Vista: A Generalizable Driving World Model with High Fidelity and Versatile Controllability2024cited by 56 here
  3. Driving into the Future: Multiview Visual Forecasting and Planning with World Model for Autonomous Driving2023cited by 55 here
  4. Epona: Autoregressive Diffusion World Model for Autonomous Driving2025cited by 39 here
  5. DriveDreamer-2: LLM-Enhanced World Models for Diverse Driving Video Generation2024cited by 35 here

Lineage

The papers this direction builds on, oldest first.

Each paper on the line cites the one before.

  1. 2022ST-P3: End-to-end Vision-based Autonomous Driving via Spatial-Temporal Feature Learningcited by 35 here
  2. 2023VAD: Vectorized Scene Representation for Efficient Autonomous Drivingcited by 60 here
  3. 2024VADv2: End-to-End Vectorized Autonomous Driving via Probabilistic Planningcited by 35 here
  4. 2024NAVSIM: Data-Driven Non-Reactive Autonomous Vehicle Simulation and Benchmarkingcited by 48 here
  5. 2024DiffusionDrive: Truncated Diffusion Model for End-to-End Autonomous Drivingcited by 48 here
  6. 2025ReCogDrive: A Reinforced Cognitive Framework for End-to-End Autonomous Drivingcited by 30 here
18 more it builds on, not in this chain
  1. 2021NuPlan: A closed-loop ML-based planning benchmark for autonomous vehiclescited by 32 here
  2. 2022TransFuser: Imitation with Transformer-Based Sensor Fusion for Autonomous Drivingcited by 39 here
  3. 2023End-to-end Autonomous Driving: Challenges and Frontierscited by 29 here
  4. 2023DriveDreamer: Towards Real-world-driven World Models for Autonomous Drivingcited by 30 here
  5. 2023GAIA-1: A Generative World Model for Autonomous Drivingcited by 64 here
  6. 2023OccWorld: Learning a 3D Occupancy World Model for Autonomous Drivingcited by 30 here
  7. 2023Driving into the Future: Multiview Visual Forecasting and Planning with World Model for Autonomous Drivingcited by 55 here
  8. 2024GenAD: Generative End-to-End Autonomous Drivingcited by 32 here
  9. 2024DriveVLM: The Convergence of Autonomous Driving and Large Vision-Language Modelscited by 24 here
  10. 2024DriveDreamer-2: LLM-Enhanced World Models for Diverse Driving Video Generationcited by 35 here
  11. 2024Vista: A Generalizable Driving World Model with High Fidelity and Versatile Controllabilitycited by 56 here
  12. 2024SparseDrive: End-to-End Autonomous Driving via Sparse Scene Representationcited by 31 here
  13. 2024Bench2Drive: Towards Multi-Ability Benchmarking of Closed-Loop End-To-End Autonomous Drivingcited by 25 here
  14. 2024Enhancing End-to-End Autonomous Driving with Latent World Modelcited by 32 here
  15. 2025End-to-End Driving with Online Trajectory Evaluation via BEV World Modelcited by 27 here
  16. 2025AutoVLA: A Vision-Language-Action Model for End-to-End Autonomous Driving with Adaptive Reasoning and Reinforcement Fine-Tuningcited by 27 here
  17. 2025Epona: Autoregressive Diffusion World Model for Autonomous Drivingcited by 39 here
  18. 2025DriveVLA-W0: World Models Amplify Data Scaling Law in Autonomous Drivingcited by 29 here

The 24 most cited of 458.

Browse

30 of 255, ranked 6–35

  1. 6GenAD: Generative End-to-End Autonomous Driving202432
  2. 7DriveDreamer: Towards Real-world-driven World Models for Autonomous Driving202330
  3. 8OccWorld: Learning a 3D Occupancy World Model for Autonomous Driving202330
  4. 9DriveVLA-W0: World Models Amplify Data Scaling Law in Autonomous Driving202529
  5. 10DriveArena: A Closed-loop Generative Simulation Platform for Autonomous Driving202424
  6. 11DrivingGPT: Unifying Driving World Modeling and Planning with Multi-modal Autoregressive Transformers202423
  7. 12GAIA-2: A Controllable Multi-View Generative World Model for Autonomous Driving202522
  8. 13DriveDreamer4D: World Models Are Effective Data Machines for 4D Driving Scene Representation202419
  9. 14FutureSightDrive: Thinking Visually with Spatio-Temporal CoT for Autonomous Driving202518
  10. 15DriveLaW:Unifying Planning and Video Generation in a Latent Driving World202517
All 30
  1. 16GenAD: Generalized Predictive Model for Autonomous Driving202416
  2. 17HUGSIM: A Real-Time, Photo-Realistic and Closed-Loop Simulator for Autonomous Driving202415
  3. 18From Forecasting to Planning: Policy World Model for Collaborative State-Action Prediction202513
  4. 19Copilot4D: Learning Unsupervised World Models for Autonomous Driving via Discrete Diffusion202313
  5. 20DriveWorld: 4D Pre-trained Scene Understanding via World Models for Autonomous Driving202412
  6. 21Unleashing Generalization of End-to-End Autonomous Driving with Controllable Long Video Generation202412
  7. 22ReconDreamer: Crafting World Models for Driving Scene Reconstruction via Online Restoration202412
  8. 23Cosmos-Drive-Dreams: Scalable Synthetic Driving Data Generation with World Foundation Models202511
  9. 24ReSim: Reliable World Simulation for Autonomous Driving202511
  10. 25Driving in the Occupancy World: Vision-Centric 4D Occupancy Forecasting and Planning via World Models for Autonomous Driving202411
  11. 26DrivingWorld: Constructing World Model for Autonomous Driving via Video GPT202411
  12. 27HERMES: A Unified Self-Driving World Model for Simultaneous 3D Scene Understanding and Generation202510
  13. 28WorldRFT: Latent World Model Planning with Reinforcement Fine-Tuning for Autonomous Driving202510
  14. 29Doe-1: Closed-Loop Autonomous Driving with Large World Model202410
  15. 30MaskGWM: A Generalizable Driving World Model with Video Mask Reconstruction20259
  16. 31Bridging Past and Future: End-to-End Autonomous Driving with Historical Prediction and Planning20259
  17. 32Enhance Sample Efficiency and Robustness of End-to-end Urban Autonomous Driving via Semantic Masked World Model20229
  18. 33DriveWorld-VLA: Unified Latent-Space World Modeling with Vision-Language-Action for Autonomous Driving20268
  19. 34VaViM and VaVAM: Autonomous Driving through Video Generative Modeling20258
  20. 35Future-Aware End-to-End Driving: Bidirectional Modeling of Trajectory Planning and Scene Evolution20258