Awesome paper list with code about End to End Autonomous Driving in NAVSIM Benchmark
100
16 commits
updated Jun 18, 2026
🚗This repository maintains a curated list of academic papers that implement and evaluate end-to-end autonomous driving models on the NavSim simulation benchmark. The goal is to track progress and compare methods within this benchmark setting.
If you find this repository useful, please giving this list a star ⭐. Feel free to share it with others!
| Method | PDMS | Venue | Date | Code | Title |
|---|---|---|---|---|---|
| BeyondDrive | 90.5 | ECCV 2026 | 19/05/2026 | Beyond Imitation: Learning Safe End-to-End Autonomous Driving from Hard Negatives | |
| CLOVER | 90.4 | arXiv | 15/05/2026 | CLOVER: Closed-Loop Value Estimation and Ranking for End-to-End Autonomous Driving Planning | |
| LTFv7 | 90.1 | ECCV 2026 | 19/05/2026 | Beyond Imitation: Learning Safe End-to-End Autonomous Driving from Hard Negatives | |
| WAM-Diff | 89.7 | arXiv | 11/12/2025 | WAM-Diff: A Masked Diffusion VLA Framework with MoE and Online Reinforcement Learning for Autonomous Driving | |
| DVGT-2 | 89.7 | CVPR2026 | 01/02/2026 | DVGT-2: Vision-Geometry-Action Model for Autonomous Driving at Scale | |
| MeanFuser | 89.5 | CVPR2026 | 13/06/2025 | MeanFuser: Fast One-Step Multi-Modal Trajectory Generation and Adaptive Reconstruction via MeanFlow for End-to-End Autonomous Driving | |
| Vega | 89.4 | arXiv | 25/03/2026 | Vega: Learning to Drive with Natural Language Instructions | |
| DiffusionDrive | 88.1 | CVPR2025 | 15/11/2024 | DiffusionDrive: Truncated Diffusion Model for End-to-End Autonomous Driving | |
| Drive-JEPA | 87.7 | arXiv | 05/01/2026 | Drive-JEPA: Video JEPA Meets Multimodal Trajectory Distillation for End-to-End Driving | |
| DriveWorld-VLA | 86.8 | ICML2026 | 06/02/2026 | DriveWorld-VLA: Unified Latent-Space World Modeling with Vision–Language–Action for Autonomous Driving | |
| Senna-2 | 86.6 | arXiv | 11/03/2026 | Senna-2: Aligning VLM and End-to-End Driving Policy for Consistent Decision Making and Planning | |
| ZTRS | 86.2 | arXiv | 14/10/2025 | ZTRS: Zero-Imitation End-to-end Autonomous Driving with Trajectory Scoring | |
| DriveVLA-W0 | 86.1 | ICLR2026 | 12/10/2025 | DriveVLA-W0: World Models Amplify Data Scaling Law in Autonomous Driving | |
| SimScale | 85.9 | CVPR2026 oral | 20/09/2025 | Learning to Drive via Real-World Simulation at Scale | |
| DiffusionDriveV2 | 85.5 | arXiv | 07/12/2025 | DiffusionDriveV2: Reinforcement Learning-Constrained Truncated Diffusion Modeling in End-to-End Autonomous Driving | |
| Curious-VLA | 85.3 | CVPR2026 Findings | 05/03/2025 | Devil is in Narrow Policy: Unleashing Exploration in Driving VLA Models | |
| WAM-Flow | 84.7 | CVPR2026 | 06/12/2025 | WAM-Flow: Parallel Coarse-to-Fine Motion Planning via Discrete Flow Matching for Autonomous Driving | |
| PRIX | 84.2 | RA-L2026 | 17/07/2025 | PRIX: Learning to Plan from Raw Pixels for End-to-End Autonomous Driving | |
| DriveSuprim | 83.1 | AAAI2026 | 10/03/2025 | DriveSuprim: Towards Precise Trajectory Selection for End-to-End Planning | |
| Latent-WAM | 89.3 | arXiv | 24/03/2026 | - | Latent-WAM: Latent World Action Modeling for End-to-End Autonomous Driving |
| ExploreVLA | 88.8 | arXiv | 02/04/2026 | coming soon | ExploreVLA: Dense World Modeling and Exploration for End-to-End Autonomous Driving |
| HAD | 88.6 | arXiv | 03/04/2026 | - | HAD: Combining Hierarchical Diffusion with Metric-Decoupled RL for End-to-End Driving |
| PaIR-Drive | 87.9 | CVPR2026 findings | 17/11/2025 | coming soon | Fine-tuning is Not Enough: A Parallel Framework for Collaborative Imitation and Reinforcement Learning in End-to-end Autonomous Driving |
| DiffRefiner | 87.9 | AAAI2026 | 17/11/2025 | coming soon | DiffRefiner: Coarse to Fine Trajectory Planning via Diffusion Refinement with Semantic Interaction for End to End Autonomous Driving |
| ELF-VLA | 87.1 | arXiv | 01/03/2026 | - | Unleashing VLA Potentials in Autonomous Driving via Explicit Learning from Failures |
| LaST-VLA | 87.1 | arXiv | 01/03/2026 | coming soon | LaST-VLA: Thinking in Latent Spatio-Temporal Space for Vision-Language-Action in Autonomous Driving |
| CdDrive | 86.4 | arXiv | 03/02/2026 | 404 | A Unified Candidate Set with Scene-Adaptive Refinement via Diffusion for End-to-End Autonomous Driving |
| SpanVLA | 86.4 | arXiv | 19/04/2026 | coming soon | SpanVLA: Efficient Action Bridging and Learning from Negative-Recovery Samples for Vision-Language-Action Model |
| The Constant Eye | 45.8 | arXiv | 12/02/2026 | - | The Constant Eye: Benchmarking and Bridging Appearance Robustness in Autonomous Driving |
| AnchDrive | 85.5 | arXiv | - | - | AnchDrive: Bootstrapping Diffusion Policies with Hybrid Trajectory Anchors for End-to-End Driving |
| Map-World | 85.0 | arXiv | 17/11/2025 | coming soon | Map-World: Masked Action planning and Path-Integral World Model for Autonomous Driving |
| SUPER-AD | 84.3 | arXiv | 22/11/2025 | - | SUPER-AD: Semantic Uncertainty-aware Planning for End-to-End Robust Autonomous Driving |
| MindDrive | 84.2 | arXiv | 04/12/2025 | - | MindDrive: An All-in-One Framework Bridging World Models and Vision-Language Model for End-to-End Autonomous Driving |
| Method | PDMS | Venue | Date | Code | Title |
|---|---|---|---|---|---|
| DrivoR | 54.6 | CVPR2026 | 05/01/2026 | Driving on Registers | |
| CLOVER | 48.3 | arXiv | 15/05/2026 | CLOVER: Closed-Loop Value Estimation and Ranking for End-to-End Autonomous Driving Planning | |
| SimScale | 48.0 | CVPR2026 oral | 20/09/2025 | Learning to Drive via Real-World Simulation at Scale | |
| ZTRS | 45.5 | arXiv | 14/10/2025 | ZTRS: Zero-Imitation End-to-end Autonomous Driving with Trajectory Scoring | |
| GTRS | 42.1 | arXiv | 06/06/2025 | Generalized Trajectory Scoring for End-to-end Multi-modal Planning | |
| RAP | 36.9 | ICLR2026 | 04/10/2026 | RAP: 3D Rasterization Augmented End-to-End Planning | |
| Mimir | 34.6 | RA-L2025 | 07/12/2025 | Mimir: Hierarchical Goal-Driven Diffusion with Uncertainty Propagation for End-to-End Autonomous Driving | |
| LEAD | 31.4 | CVPR2026 | 20/12/2025 | LEAD: Minimizing Learner-Expert Asymmetry in End-to-End Driving | |
| MindDrive | 30.5 | arXiv | 04/12/2025 | - | MindDrive: An All-in-One Framework Bridging World Models and Vision-Language Model for End-to-End Autonomous Driving |
| LTFv6 | 28.3 | CVPR2026 | 20/12/2025 | LEAD: Minimizing Learner-Expert Asymmetry in End-to-End Driving | |
| The Constant Eye | 46.7 | arXiv | 12/02/2026 | - | The Constant Eye: Benchmarking and Bridging Appearance Robustness in Autonomous Driving |
| SpanVLA | 40.1 | arXiv | 19/04/2026 | coming soon | SpanVLA: Efficient Action Bridging and Learning from Negative-Recovery Samples for Vision-Language-Action Model |
| EponaV2 | 36.1 | arXiv | 14/05/2026 | coming soon | EponaV2: Driving World Model with Comprehensive Future Reasoning |
| UniUncer | 28.7 | arXiv | 07/03/2026 | - | UniUncer: Unified Dynamic–Static Uncertainty for End-to-End Driving |
| Method | PDMS | Venue | Date | Code | Title |
|---|---|---|---|---|---|
| CLOVER | 94.5 | arXiv | 15/05/2026 | CLOVER: Closed-Loop Value Estimation and Ranking for End-to-End Autonomous Driving Planning | |
| RAP | 93.8 | ICLR2026 | 04/10/2025 | RAP: 3D Rasterization Augmented End-to-End Planning | |
| DriveSuprim | 93.5 | AAAI2026 | 10/03/2025 | DriveSuprim: Towards Precise Trajectory Selection for End-to-End Planning | |
| Drive-JEPA | 93.3 | arXiv | 05/01/2026 | Drive-JEPA: Video JEPA Meets Multimodal Trajectory Distillation for End-to-End Driving | |
| DrivoR | 93.1 | CVPR2026 | 05/01/2026 | Driving on Registers | |
| SparseDriveV2 | 92.0 | arXiv | 29/03/2026 | SparseDriveV2: Scoring is All You Need for End-to-End Autonomous Driving | |
| iPad | 91.7 | arXiv | 15/05/2025 | iPad: Iterative Proposal-centric End-to-End Autonomous Driving | |
| DriveWorld-VLA | 86.8 | ICML2026 | 06/02/2026 | DriveWorld-VLA: Unified Latent-Space World Modeling with Vision–Language–Action for Autonomous Driving | |
| DiffusionDriveV2 | 91.2 | arXiv | 07/12/2025 | DiffusionDriveV2: Reinforcement Learning-Constrained Truncated Diffusion Modeling in End-to-End Autonomous Driving | |
| SGDrive | 91.1 | CVPR2026 | 05/01/2026 | SGDrive: Scene-to-Goal Hierarchical World Cognition for Autonomous Driving | |
| WAM-Diff | 91.0 | arXiv | score | WAM-Diff: A Masked Diffusion VLA Framework with MoE and Online Reinforcement Learning for Autonomous Driving | |
| ReCogDrive | 90.8 | ICLR2026 | 08/06/2025 | ReCogDrive: A Reinforced Cognitive Framework for End-to-End Autonomous Driving | |
| SeerDrive | 90.7 | NeurIPS2025 | 11/10/2025 | Future-Aware End-to-End Driving: Bidirectional Modeling of Trajectory Planning and Scene Evolution | |
| GoalFlow | 90.3 | CVPR2025 | 05/03/2025 | GoalFlow: Goal-Driven Flow Matching for Multimodal Trajectories Generation in End-to-End Autonomous Driving | |
| WAM-Flow | 90.3 | CVPR2026 | 06/12/2025 | WAM-Flow: Parallel Coarse-to-Fine Motion Planning via Discrete Flow Matching for Autonomous Driving | |
| Curious-VLA | 90.3 | CVPR2026 Findings | 05/03/2025 | Devil is in Narrow Policy: Unleashing Exploration in Driving VLA Models | |
| DVGT-2 | 90.3 | CVPR2026 | 01/04/2026 | DVGT-2: Vision-Geometry-Action Model for Autonomous Driving at Scale | |
| BeyondDrive | 90.3 | arXiv | 19/05/2026 | Beyond Imitation: Learning Safe End-to-End Autonomous Driving from Hard Negatives | |
| DriveVLA-W0 | 90.2 | ICLR2026 | 12/10/2025 | DriveVLA-W0: World Models Amplify Data Scaling Law in Autonomous Driving | |
| VeteranAD | 90.2 | AAAI2026 | 11/08/2025 | Perception in Plan: Coupled Perception and Planning for End-to-End Autonomous Driving | |
| DeMo++ | 89.9 | arXiv | 17/07/2025 | DeMo++: Motion Decoupling for Autonomous Driving | |
| LTFv7 | 89.7 | arXiv | 19/05/2026 | Beyond Imitation: Learning Safe End-to-End Autonomous Driving from Hard Negatives | |
| Uni-World VLA | 89.4 | arXiv | 27/03/2026 | Uni-World VLA: Interleaved World Modeling and Planning for Autonomous Driving | |
| Mimir | 89.3 | RA-L2025 | 07/12/2025 | Mimir: Hierarchical Goal-Driven Diffusion with Uncertainty Propagation for End-to-End Autonomous Driving | |
| AutoVLA | 89.1 | NeurIPS2025 | 13/06/2025 | AutoVLA: A Vision-Language-Action Model for End-to-End Autonomous Driving with Adaptive Reasoning and Reinforcement Fine-Tuning | |
| DriveLaW | 89.1 | CVPR2026 | 23/12/2025 | DriveLaW:Unifying Planning and Video Generation in a Latent Driving World | |
| MeanFuser | 89.0 | CVPR2026 | 13/06/2025 | MeanFuser: Fast One-Step Multi-Modal Trajectory Generation and Adaptive Reconstruction via MeanFlow for End-to-End Autonomous Driving | |
| VGGDrive | 88.8 | CVPR2026 | 13/06/2025 | VGGDrive: Empowering Vision-Language Models with Cross-View Geometric Grounding for Autonomous Driving | |
| Hydra-NeXt | 88.6 | ICCV2025 | 12/03/2025 | Hydra-NeXt: Robust Closed-Loop Driving with Open-Loop Training | |
| WoTE | 88.3 | ICCV2025 | 01/04/2025 | End-to-End Driving with Online Trajectory Evaluation via BEV World Model | |
| DIVER | 88.3 | arXiv | 04/07/2025 | DIVER: Reinforced Diffusion Breaks Imitation Bottlenecks in End-to-End Autonomous Driving | |
| DiffusionDrive | 88.1 | CVPR2025 | 15/11/2024 | DiffusionDrive: Truncated Diffusion Model for End-to-End Autonomous Driving | |
| PWM | 88.1 | NeurIPS2025 | 19/10/2024 | From Forecasting to Planning: Policy World Model for Collaborative State-Action Prediction | |
| BridgeDrive | 88.0 | ICLR2026 | 23/09/2024 | BridgeDrive: Diffusion Bridge Policy for Closed-Loop Trajectory Planning in Autonomous Driving | |
| Vega | 87.9 | arXiv | 25/03/20264 | Vega: Learning to Drive with Natural Language Instructions | |
| PRIX | 87.8 | RA-L2026 | 17/07/2025 | PRIX: Learning to Plan from Raw Pixels for End-to-End Autonomous Driving | |
| ImagiDrive | 87.4 | ICRA2026 | 11/08/2025 | ImagiDrive: A Unified Imagination-and-Planning Framework for Autonomous Driving | |
| DistillDrive | 86.2 | ICCV2025 | 05/08/2025 | DistillDrive: End-to-End Multi-Mode Autonomous Driving Distillation by Isomorphic Hetero-Source Planning Model | |
| LEAD | 86.4 | CVPR2026 | 20/12/2025 | LEAD: Minimizing Learner-Expert Asymmetry in End-to-End Driving | |
| Epona | 86.2 | ICCV2025 | 24/06/2025 | Epona: Autoregressive Diffusion World Model for Autonomous Driving | |
| LTFv6 | 85.4 | CVPR2026 | 20/12/2025 | LEAD: Minimizing Learner-Expert Asymmetry in End-to-End Driving | |
| World4Drive | 85.1 | ICCV2025 | 01/07/2025 | World4Drive: End-to-End Autonomous Driving via Intention-aware Physical Latent World Model | |
| FSDrive | 85.1 | ICCV2025 | 17/05/2025 | FutureSightDrive: Thinking Visually with Spatio-Temporal CoT for Autonomous Driving | |
| LAW | 84.6 | ICLR2025 | 08/06/2024 | Enhancing End-to-End Autonomous Driving with Latent World Model | |
| OneDrive | 86.8 | arXiv | 17/04/2026 | OneDrive: Unified Multi-Paradigm Driving with Vision-Language-Action Models | |
| UniVLA | 81.7 | ICLR2026 | 19/06/2025 | Unified Vision-Language-Action Model | |
| EvaDrive | 94.9 | arXiv | 09/08/2025 | - | EvaDrive: Evolutionary Adversarial Policy Optimization for End-to-End Autonomous Driving |
| TransDiffuser | 94.9 | arXiv | 09/05/2025 | - | TransDiffuser: Diverse Trajectory Generation with Decorrelated Multi-modal Representation for End-to-end Autonomous Driving |
| DiffE2E | 92.7 | NeurIPS2025 | 19/05/2025 | coming soon | DiffE2E: Rethinking End-to-End Driving with a Hybrid Action Diffusion and Supervised Policy |
| NaviHydra | 92.7 | arXiv | 10/12/2025 | - | NaviHydra: Controllable Navigation-guided End-to-end Autonomous Driving with Hydra-distillation |
| Centaur | 92.6 | arXiv | 11/03/2025 | - | Centaur: Robust End-to-End Autonomous Driving with Test-Time Training |
| HiPro-AD | 92.6 | - | - | - | HiPro-AD: Sparse Trajectory Transformer for End-to-End Autonomous Driving with Hybrid Spatiotemporal Attention |
| LatentVLA | 92.4 | arXiv | 05/01/2026 | - | LatentVLA: Efficient Vision-Language Models for Autonomous Driving via Latent Action Prediction |
| AD-R1 | 91.9 | arXiv | 20/11/2025 | - | AD-R1: Closed-Loop Reinforcement Learning for End-to-End Autonomous Driving with Impartial World Models |
| DriveFine | 91.8 | arXiv | 14/02/2025 | - | DriveFine: Refining-Augmented Masked Diffusion VLA for Precise and Robust Driving |
| DynVLA | 91.7 | arXiv | 11/03/2026 | - | DynVLA: Learning World Dynamics for Action Reasoning in Autonomous Driving |
| R2SE | 91.6 | TPAMI2026 | 09/06/2025 | - | Reinforced Refinement with Self-Aware Expansion for End-to-End Autonomous Driving |
| SafeDrive | 91.6 | CVPR2026 | 18/02/2026 | coming soon | SafeDrive: Fine-Grained Safety Reasoning for End-to-End Driving in a Sparse World |
| LaST-VLA | 91.3 | arXiv | 01/03/2026 | coming soon | LaST-VLA: Thinking in Latent Spatio-Temporal Space for Vision-Language-Action in Autonomous Driving |
| RaWMPC | 91.3 | arXiv | 23/02/2026 | - | Risk-Aware World Model Predictive Control for Generalizable End-to-End Autonomous Driving |
| PaIR-Drive | 91.2 | CVPR2026 findings | 17/11/2025 | coming soon | Fine-tuning is Not Enough: A Parallel Framework for Collaborative Imitation and Reinforcement Learning in End-to-end Autonomous Driving |
| ReflectDrive | 91.1 | arXiv | 20/09/2025 | - | Discrete Diffusion for Reflective Vision-Language-Action Models in Autonomous Driving |
| ReflectDrive-2 | 91.0 | arXiv | 04/05/2026 | - | ReflectDrive-2: Reinforcement-Learning-Aligned Self-Editing for Discrete Diffusion Driving |
| Hydra-MDP++ | 91.0 | arXiv | 12/03/2025 | - | Hydra-MDP++: Advancing End-to-End Driving via Expert-Guided Hydra-Distillation |
| ELF-VLA | 91.0 | arXiv | 01/03/2026 | - | Unleashing VLA Potentials in Autonomous Driving via Explicit Learning from Failures |
| DriveVA | 90.9 | arXiv | 03/03/2026 | - | DriveVA: Video Action Models are Zero-Shot Drivers |
| LADY | 90.9 | arXiv | 15/12/2025 | - | LADY: Linear Attention for Autonomous Driving Efficiency without Transformers |
| DiffRefiner | 90.7 | AAAI2026 | 17/11/2025 | coming soon | DiffRefiner: Coarse to Fine Trajectory Planning via Diffusion Refinement with Semantic Interaction for End to End Autonomous Driving |
| FutureX | 90.6 | arXiv | 11/12/2025 | coming soon | FutureX: Enhance End-to-End Autonomous Driving via Latent Chain-of-Thought World Model |
| UniDWM | 90.6 | arXiv | 01/02/2026 | 404 | UniDWM: Towards a Unified Driving World Model via Multifaceted Representation Learning |
| ResAD | 90.6 | arXiv | 08/10/2025 | coming soon | ResAD: Normalized Residual Trajectory Modeling for End-to-End Autonomous Driving |
| ExploreVLA | 90.4 | arXiv | 02/04/2026 | coming soon | ExploreVLA: Dense World Modeling and Exploration for End-to-End Autonomous Driving |
| EponaV2 | 36.1 | arXiv | - | coming soon | EponaV2: Driving World Model with Comprehensive Future Reasoning |
| SpanVLA | 90.3 | arXiv | 19/04/2026 | coming soon | SpanVLA: Efficient Action Bridging and Learning from Negative-Recovery Samples for Vision-Language-Action Model |
| AdaThinkDrive | 90.3 | CVPR2025 | 13/09/2025 | - | AdaThinkDrive: Adaptive Thinking via Reinforcement Learning for Autonomous Driving |
| HAD | 90.2 | arXiv | 03/04/2026 | - | HAD: Combining Hierarchical Diffusion with Metric-Decoupled RL for End-to-End Driving |
| FeaXDrive | 90.0 | NeurIPS2025 | 12/04/2026 | - | FeaXDrive: Feasibility-aware Trajectory-Centric Diffusion Planning for End-to-End Autonomous Driving |
| DriveDPO | 90.0 | NeurIPS2025 | 17/09/2025 | - | DriveDPO: Policy Learning via Safety DPO For End-to-End Autonomous Driving |
| - | 89.8 | arXiv | 20/09/2025 | - | Autoregressive End-to-End Planning with Time-Invariant Spatial Alignment and Multi-Objective Policy Refinement |
| CdDrive | 89.2 | arXiv | 03/02/2026 | 404 | A Unified Candidate Set with Scene-Adaptive Refinement via Diffusion for End-to-End Autonomous Driving |
| MindDrive | 88.9 | arXiv | 04/12/2025 | - | MindDrive: An All-in-One Framework Bridging World Models and Vision-Language Model for End-to-End Autonomous Driving |
| GMF-Drive | 88.9 | arXiv | 06/08/2025 | - | GMF-Drive: Gated Mamba Fusion with Spatial-Aware BEV Representation for End-to-End Autonomous Driving |
| Map-World | 88.8 | arXiv | 20/11/2025 | - | Map-World: Masked Action planning and Path-Integral World Model for Autonomous Driving |
| READ | 88.8 | - | 13/09/2025 | - | READ: End-to-End Autonomous Driving Made Safer with Efficient Reinforcement Learning |
| DynFlowDrive | 88.7 | - | 19/03/2026 | coming soon | DynFlowDrive: Flow-Based Dynamic World Modeling for Autonomous Driving |
| DreamerAD | 88.7 | - | 24/03/2026 | - | DreamerAD: Efficient Reinforcement Learning via Latent World Model for Autonomous Driving |
| TrajDiff | 88.5 | arXiv | 01/12/2025 | coming soon | TrajDiff: End-to-end Autonomous Driving without Perception Annotation |
| WorldRFT | 87.8 | AAAI2026 | 19/12/2025 | coming soon | WorldRFT: Latent World Model Planning with Reinforcement Fine-Tuning for Autonomous Driving |
| SUPER-AD | 87.7 | arXiv | 22/11/2025 | - | SUPER-AD: Semantic Uncertainty-aware Planning for End-to-End Robust Autonomous Driving |
| TrajHF | 87.6 | arXiv | 10/03/2025 | - | Learning Personalized Driving Styles via Reinforcement Learning from Human Feedback |
| CausalVAD | 87.2 | arXiv | 18/03/2026 | - | CausalVAD: De-confounding End-to-End Autonomous Driving via Causal Intervention |
| DAP | 87.2 | arXiv | 13/11/2025 | - | DAP: A Discrete-token Autoregressive Planner for Autonomous Driving |
| ARTEMIS | 87.0 | arXiv | 19/04/2025 | coming soon | ARTEMIS: Autoregressive End-to-End Trajectory Planning with Mixture of Experts for Autonomous Driving |
| ProDrive | 86.6 | arXiv | 25/04/2026 | - | ProDrive: Proactive Planning for Autonomous Driving via Ego-Environment Co-Evolution |
| NoRD | 85.6 | arXiv | 21/02/2026 | - | NoRD: A Data-Efficient Vision-Language-Action Model that Drives without Reasoning |
| LFG | 85.2 | arXiv | 22/02/2026 | - | Learning to Drive is a Free Gift: Large-Scale Label-Free Autonomy Pretraining from Unposed In-The-Wild Videos |
| DriveX | 84.5 | ICCV2025 | 19/05/2025 | - | DriveX: Omni Scene Modeling for Learning Generalizable World Knowledge in Autonomous Driving |
| DISK | 83.6 | arXiv | 01/02/2026 | - | DISK: Dynamic Inference SKipping for World Models |
| DrivingGPT | 82.4 | ICCV2025 | 18/12/2024 | 404 | DrivingGPT: Unifying Driving World Modeling and Planning with Multi-modal Autoregressive Transformers |
Awesome paper list with code about End to End Autonomous Driving in NAVSIM Benchmark
100
16 commits
updated Jun 18, 2026
🚗This repository maintains a curated list of academic papers that implement and evaluate end-to-end autonomous driving models on the NavSim simulation benchmark. The goal is to track progress and compare methods within this benchmark setting.
If you find this repository useful, please giving this list a star ⭐. Feel free to share it with others!
| Method | PDMS | Venue | Date | Code | Title |
|---|---|---|---|---|---|
| BeyondDrive | 90.5 | ECCV 2026 | 19/05/2026 | Beyond Imitation: Learning Safe End-to-End Autonomous Driving from Hard Negatives | |
| CLOVER | 90.4 | arXiv | 15/05/2026 | CLOVER: Closed-Loop Value Estimation and Ranking for End-to-End Autonomous Driving Planning | |
| LTFv7 | 90.1 | ECCV 2026 | 19/05/2026 | Beyond Imitation: Learning Safe End-to-End Autonomous Driving from Hard Negatives | |
| WAM-Diff | 89.7 | arXiv | 11/12/2025 | WAM-Diff: A Masked Diffusion VLA Framework with MoE and Online Reinforcement Learning for Autonomous Driving | |
| DVGT-2 | 89.7 | CVPR2026 | 01/02/2026 | DVGT-2: Vision-Geometry-Action Model for Autonomous Driving at Scale | |
| MeanFuser | 89.5 | CVPR2026 | 13/06/2025 | MeanFuser: Fast One-Step Multi-Modal Trajectory Generation and Adaptive Reconstruction via MeanFlow for End-to-End Autonomous Driving | |
| Vega | 89.4 | arXiv | 25/03/2026 | Vega: Learning to Drive with Natural Language Instructions | |
| DiffusionDrive | 88.1 | CVPR2025 | 15/11/2024 | DiffusionDrive: Truncated Diffusion Model for End-to-End Autonomous Driving | |
| Drive-JEPA | 87.7 | arXiv | 05/01/2026 | Drive-JEPA: Video JEPA Meets Multimodal Trajectory Distillation for End-to-End Driving | |
| DriveWorld-VLA | 86.8 | ICML2026 | 06/02/2026 | DriveWorld-VLA: Unified Latent-Space World Modeling with Vision–Language–Action for Autonomous Driving | |
| Senna-2 | 86.6 | arXiv | 11/03/2026 | Senna-2: Aligning VLM and End-to-End Driving Policy for Consistent Decision Making and Planning | |
| ZTRS | 86.2 | arXiv | 14/10/2025 | ZTRS: Zero-Imitation End-to-end Autonomous Driving with Trajectory Scoring | |
| DriveVLA-W0 | 86.1 | ICLR2026 | 12/10/2025 | DriveVLA-W0: World Models Amplify Data Scaling Law in Autonomous Driving | |
| SimScale | 85.9 | CVPR2026 oral | 20/09/2025 | Learning to Drive via Real-World Simulation at Scale | |
| DiffusionDriveV2 | 85.5 | arXiv | 07/12/2025 | DiffusionDriveV2: Reinforcement Learning-Constrained Truncated Diffusion Modeling in End-to-End Autonomous Driving | |
| Curious-VLA | 85.3 | CVPR2026 Findings | 05/03/2025 | Devil is in Narrow Policy: Unleashing Exploration in Driving VLA Models | |
| WAM-Flow | 84.7 | CVPR2026 | 06/12/2025 | WAM-Flow: Parallel Coarse-to-Fine Motion Planning via Discrete Flow Matching for Autonomous Driving | |
| PRIX | 84.2 | RA-L2026 | 17/07/2025 | PRIX: Learning to Plan from Raw Pixels for End-to-End Autonomous Driving | |
| DriveSuprim | 83.1 | AAAI2026 | 10/03/2025 | DriveSuprim: Towards Precise Trajectory Selection for End-to-End Planning | |
| Latent-WAM | 89.3 | arXiv | 24/03/2026 | - | Latent-WAM: Latent World Action Modeling for End-to-End Autonomous Driving |
| ExploreVLA | 88.8 | arXiv | 02/04/2026 | coming soon | ExploreVLA: Dense World Modeling and Exploration for End-to-End Autonomous Driving |
| HAD | 88.6 | arXiv | 03/04/2026 | - | HAD: Combining Hierarchical Diffusion with Metric-Decoupled RL for End-to-End Driving |
| PaIR-Drive | 87.9 | CVPR2026 findings | 17/11/2025 | coming soon | Fine-tuning is Not Enough: A Parallel Framework for Collaborative Imitation and Reinforcement Learning in End-to-end Autonomous Driving |
| DiffRefiner | 87.9 | AAAI2026 | 17/11/2025 | coming soon | DiffRefiner: Coarse to Fine Trajectory Planning via Diffusion Refinement with Semantic Interaction for End to End Autonomous Driving |
| ELF-VLA | 87.1 | arXiv | 01/03/2026 | - | Unleashing VLA Potentials in Autonomous Driving via Explicit Learning from Failures |
| LaST-VLA | 87.1 | arXiv | 01/03/2026 | coming soon | LaST-VLA: Thinking in Latent Spatio-Temporal Space for Vision-Language-Action in Autonomous Driving |
| CdDrive | 86.4 | arXiv | 03/02/2026 | 404 | A Unified Candidate Set with Scene-Adaptive Refinement via Diffusion for End-to-End Autonomous Driving |
| SpanVLA | 86.4 | arXiv | 19/04/2026 | coming soon | SpanVLA: Efficient Action Bridging and Learning from Negative-Recovery Samples for Vision-Language-Action Model |
| The Constant Eye | 45.8 | arXiv | 12/02/2026 | - | The Constant Eye: Benchmarking and Bridging Appearance Robustness in Autonomous Driving |
| AnchDrive | 85.5 | arXiv | - | - | AnchDrive: Bootstrapping Diffusion Policies with Hybrid Trajectory Anchors for End-to-End Driving |
| Map-World | 85.0 | arXiv | 17/11/2025 | coming soon | Map-World: Masked Action planning and Path-Integral World Model for Autonomous Driving |
| SUPER-AD | 84.3 | arXiv | 22/11/2025 | - | SUPER-AD: Semantic Uncertainty-aware Planning for End-to-End Robust Autonomous Driving |
| MindDrive | 84.2 | arXiv | 04/12/2025 | - | MindDrive: An All-in-One Framework Bridging World Models and Vision-Language Model for End-to-End Autonomous Driving |
| Method | PDMS | Venue | Date | Code | Title |
|---|---|---|---|---|---|
| DrivoR | 54.6 | CVPR2026 | 05/01/2026 | Driving on Registers | |
| CLOVER | 48.3 | arXiv | 15/05/2026 | CLOVER: Closed-Loop Value Estimation and Ranking for End-to-End Autonomous Driving Planning | |
| SimScale | 48.0 | CVPR2026 oral | 20/09/2025 | Learning to Drive via Real-World Simulation at Scale | |
| ZTRS | 45.5 | arXiv | 14/10/2025 | ZTRS: Zero-Imitation End-to-end Autonomous Driving with Trajectory Scoring | |
| GTRS | 42.1 | arXiv | 06/06/2025 | Generalized Trajectory Scoring for End-to-end Multi-modal Planning | |
| RAP | 36.9 | ICLR2026 | 04/10/2026 | RAP: 3D Rasterization Augmented End-to-End Planning | |
| Mimir | 34.6 | RA-L2025 | 07/12/2025 | Mimir: Hierarchical Goal-Driven Diffusion with Uncertainty Propagation for End-to-End Autonomous Driving | |
| LEAD | 31.4 | CVPR2026 | 20/12/2025 | LEAD: Minimizing Learner-Expert Asymmetry in End-to-End Driving | |
| MindDrive | 30.5 | arXiv | 04/12/2025 | - | MindDrive: An All-in-One Framework Bridging World Models and Vision-Language Model for End-to-End Autonomous Driving |
| LTFv6 | 28.3 | CVPR2026 | 20/12/2025 | LEAD: Minimizing Learner-Expert Asymmetry in End-to-End Driving | |
| The Constant Eye | 46.7 | arXiv | 12/02/2026 | - | The Constant Eye: Benchmarking and Bridging Appearance Robustness in Autonomous Driving |
| SpanVLA | 40.1 | arXiv | 19/04/2026 | coming soon | SpanVLA: Efficient Action Bridging and Learning from Negative-Recovery Samples for Vision-Language-Action Model |
| EponaV2 | 36.1 | arXiv | 14/05/2026 | coming soon | EponaV2: Driving World Model with Comprehensive Future Reasoning |
| UniUncer | 28.7 | arXiv | 07/03/2026 | - | UniUncer: Unified Dynamic–Static Uncertainty for End-to-End Driving |
| Method | PDMS | Venue | Date | Code | Title |
|---|---|---|---|---|---|
| CLOVER | 94.5 | arXiv | 15/05/2026 | CLOVER: Closed-Loop Value Estimation and Ranking for End-to-End Autonomous Driving Planning | |
| RAP | 93.8 | ICLR2026 | 04/10/2025 | RAP: 3D Rasterization Augmented End-to-End Planning | |
| DriveSuprim | 93.5 | AAAI2026 | 10/03/2025 | DriveSuprim: Towards Precise Trajectory Selection for End-to-End Planning | |
| Drive-JEPA | 93.3 | arXiv | 05/01/2026 | Drive-JEPA: Video JEPA Meets Multimodal Trajectory Distillation for End-to-End Driving | |
| DrivoR | 93.1 | CVPR2026 | 05/01/2026 | Driving on Registers | |
| SparseDriveV2 | 92.0 | arXiv | 29/03/2026 | SparseDriveV2: Scoring is All You Need for End-to-End Autonomous Driving | |
| iPad | 91.7 | arXiv | 15/05/2025 | iPad: Iterative Proposal-centric End-to-End Autonomous Driving | |
| DriveWorld-VLA | 86.8 | ICML2026 | 06/02/2026 | DriveWorld-VLA: Unified Latent-Space World Modeling with Vision–Language–Action for Autonomous Driving | |
| DiffusionDriveV2 | 91.2 | arXiv | 07/12/2025 | DiffusionDriveV2: Reinforcement Learning-Constrained Truncated Diffusion Modeling in End-to-End Autonomous Driving | |
| SGDrive | 91.1 | CVPR2026 | 05/01/2026 | SGDrive: Scene-to-Goal Hierarchical World Cognition for Autonomous Driving | |
| WAM-Diff | 91.0 | arXiv | score | WAM-Diff: A Masked Diffusion VLA Framework with MoE and Online Reinforcement Learning for Autonomous Driving | |
| ReCogDrive | 90.8 | ICLR2026 | 08/06/2025 | ReCogDrive: A Reinforced Cognitive Framework for End-to-End Autonomous Driving | |
| SeerDrive | 90.7 | NeurIPS2025 | 11/10/2025 | Future-Aware End-to-End Driving: Bidirectional Modeling of Trajectory Planning and Scene Evolution | |
| GoalFlow | 90.3 | CVPR2025 | 05/03/2025 | GoalFlow: Goal-Driven Flow Matching for Multimodal Trajectories Generation in End-to-End Autonomous Driving | |
| WAM-Flow | 90.3 | CVPR2026 | 06/12/2025 | WAM-Flow: Parallel Coarse-to-Fine Motion Planning via Discrete Flow Matching for Autonomous Driving | |
| Curious-VLA | 90.3 | CVPR2026 Findings | 05/03/2025 | Devil is in Narrow Policy: Unleashing Exploration in Driving VLA Models | |
| DVGT-2 | 90.3 | CVPR2026 | 01/04/2026 | DVGT-2: Vision-Geometry-Action Model for Autonomous Driving at Scale | |
| BeyondDrive | 90.3 | arXiv | 19/05/2026 | Beyond Imitation: Learning Safe End-to-End Autonomous Driving from Hard Negatives | |
| DriveVLA-W0 | 90.2 | ICLR2026 | 12/10/2025 | DriveVLA-W0: World Models Amplify Data Scaling Law in Autonomous Driving | |
| VeteranAD | 90.2 | AAAI2026 | 11/08/2025 | Perception in Plan: Coupled Perception and Planning for End-to-End Autonomous Driving | |
| DeMo++ | 89.9 | arXiv | 17/07/2025 | DeMo++: Motion Decoupling for Autonomous Driving | |
| LTFv7 | 89.7 | arXiv | 19/05/2026 | Beyond Imitation: Learning Safe End-to-End Autonomous Driving from Hard Negatives | |
| Uni-World VLA | 89.4 | arXiv | 27/03/2026 | Uni-World VLA: Interleaved World Modeling and Planning for Autonomous Driving | |
| Mimir | 89.3 | RA-L2025 | 07/12/2025 | Mimir: Hierarchical Goal-Driven Diffusion with Uncertainty Propagation for End-to-End Autonomous Driving | |
| AutoVLA | 89.1 | NeurIPS2025 | 13/06/2025 | AutoVLA: A Vision-Language-Action Model for End-to-End Autonomous Driving with Adaptive Reasoning and Reinforcement Fine-Tuning | |
| DriveLaW | 89.1 | CVPR2026 | 23/12/2025 | DriveLaW:Unifying Planning and Video Generation in a Latent Driving World | |
| MeanFuser | 89.0 | CVPR2026 | 13/06/2025 | MeanFuser: Fast One-Step Multi-Modal Trajectory Generation and Adaptive Reconstruction via MeanFlow for End-to-End Autonomous Driving | |
| VGGDrive | 88.8 | CVPR2026 | 13/06/2025 | VGGDrive: Empowering Vision-Language Models with Cross-View Geometric Grounding for Autonomous Driving | |
| Hydra-NeXt | 88.6 | ICCV2025 | 12/03/2025 | Hydra-NeXt: Robust Closed-Loop Driving with Open-Loop Training | |
| WoTE | 88.3 | ICCV2025 | 01/04/2025 | End-to-End Driving with Online Trajectory Evaluation via BEV World Model | |
| DIVER | 88.3 | arXiv | 04/07/2025 | DIVER: Reinforced Diffusion Breaks Imitation Bottlenecks in End-to-End Autonomous Driving | |
| DiffusionDrive | 88.1 | CVPR2025 | 15/11/2024 | DiffusionDrive: Truncated Diffusion Model for End-to-End Autonomous Driving | |
| PWM | 88.1 | NeurIPS2025 | 19/10/2024 | From Forecasting to Planning: Policy World Model for Collaborative State-Action Prediction | |
| BridgeDrive | 88.0 | ICLR2026 | 23/09/2024 | BridgeDrive: Diffusion Bridge Policy for Closed-Loop Trajectory Planning in Autonomous Driving | |
| Vega | 87.9 | arXiv | 25/03/20264 | Vega: Learning to Drive with Natural Language Instructions | |
| PRIX | 87.8 | RA-L2026 | 17/07/2025 | PRIX: Learning to Plan from Raw Pixels for End-to-End Autonomous Driving | |
| ImagiDrive | 87.4 | ICRA2026 | 11/08/2025 | ImagiDrive: A Unified Imagination-and-Planning Framework for Autonomous Driving | |
| DistillDrive | 86.2 | ICCV2025 | 05/08/2025 | DistillDrive: End-to-End Multi-Mode Autonomous Driving Distillation by Isomorphic Hetero-Source Planning Model | |
| LEAD | 86.4 | CVPR2026 | 20/12/2025 | LEAD: Minimizing Learner-Expert Asymmetry in End-to-End Driving | |
| Epona | 86.2 | ICCV2025 | 24/06/2025 | Epona: Autoregressive Diffusion World Model for Autonomous Driving | |
| LTFv6 | 85.4 | CVPR2026 | 20/12/2025 | LEAD: Minimizing Learner-Expert Asymmetry in End-to-End Driving | |
| World4Drive | 85.1 | ICCV2025 | 01/07/2025 | World4Drive: End-to-End Autonomous Driving via Intention-aware Physical Latent World Model | |
| FSDrive | 85.1 | ICCV2025 | 17/05/2025 | FutureSightDrive: Thinking Visually with Spatio-Temporal CoT for Autonomous Driving | |
| LAW | 84.6 | ICLR2025 | 08/06/2024 | Enhancing End-to-End Autonomous Driving with Latent World Model | |
| OneDrive | 86.8 | arXiv | 17/04/2026 | OneDrive: Unified Multi-Paradigm Driving with Vision-Language-Action Models | |
| UniVLA | 81.7 | ICLR2026 | 19/06/2025 | Unified Vision-Language-Action Model | |
| EvaDrive | 94.9 | arXiv | 09/08/2025 | - | EvaDrive: Evolutionary Adversarial Policy Optimization for End-to-End Autonomous Driving |
| TransDiffuser | 94.9 | arXiv | 09/05/2025 | - | TransDiffuser: Diverse Trajectory Generation with Decorrelated Multi-modal Representation for End-to-end Autonomous Driving |
| DiffE2E | 92.7 | NeurIPS2025 | 19/05/2025 | coming soon | DiffE2E: Rethinking End-to-End Driving with a Hybrid Action Diffusion and Supervised Policy |
| NaviHydra | 92.7 | arXiv | 10/12/2025 | - | NaviHydra: Controllable Navigation-guided End-to-end Autonomous Driving with Hydra-distillation |
| Centaur | 92.6 | arXiv | 11/03/2025 | - | Centaur: Robust End-to-End Autonomous Driving with Test-Time Training |
| HiPro-AD | 92.6 | - | - | - | HiPro-AD: Sparse Trajectory Transformer for End-to-End Autonomous Driving with Hybrid Spatiotemporal Attention |
| LatentVLA | 92.4 | arXiv | 05/01/2026 | - | LatentVLA: Efficient Vision-Language Models for Autonomous Driving via Latent Action Prediction |
| AD-R1 | 91.9 | arXiv | 20/11/2025 | - | AD-R1: Closed-Loop Reinforcement Learning for End-to-End Autonomous Driving with Impartial World Models |
| DriveFine | 91.8 | arXiv | 14/02/2025 | - | DriveFine: Refining-Augmented Masked Diffusion VLA for Precise and Robust Driving |
| DynVLA | 91.7 | arXiv | 11/03/2026 | - | DynVLA: Learning World Dynamics for Action Reasoning in Autonomous Driving |
| R2SE | 91.6 | TPAMI2026 | 09/06/2025 | - | Reinforced Refinement with Self-Aware Expansion for End-to-End Autonomous Driving |
| SafeDrive | 91.6 | CVPR2026 | 18/02/2026 | coming soon | SafeDrive: Fine-Grained Safety Reasoning for End-to-End Driving in a Sparse World |
| LaST-VLA | 91.3 | arXiv | 01/03/2026 | coming soon | LaST-VLA: Thinking in Latent Spatio-Temporal Space for Vision-Language-Action in Autonomous Driving |
| RaWMPC | 91.3 | arXiv | 23/02/2026 | - | Risk-Aware World Model Predictive Control for Generalizable End-to-End Autonomous Driving |
| PaIR-Drive | 91.2 | CVPR2026 findings | 17/11/2025 | coming soon | Fine-tuning is Not Enough: A Parallel Framework for Collaborative Imitation and Reinforcement Learning in End-to-end Autonomous Driving |
| ReflectDrive | 91.1 | arXiv | 20/09/2025 | - | Discrete Diffusion for Reflective Vision-Language-Action Models in Autonomous Driving |
| ReflectDrive-2 | 91.0 | arXiv | 04/05/2026 | - | ReflectDrive-2: Reinforcement-Learning-Aligned Self-Editing for Discrete Diffusion Driving |
| Hydra-MDP++ | 91.0 | arXiv | 12/03/2025 | - | Hydra-MDP++: Advancing End-to-End Driving via Expert-Guided Hydra-Distillation |
| ELF-VLA | 91.0 | arXiv | 01/03/2026 | - | Unleashing VLA Potentials in Autonomous Driving via Explicit Learning from Failures |
| DriveVA | 90.9 | arXiv | 03/03/2026 | - | DriveVA: Video Action Models are Zero-Shot Drivers |
| LADY | 90.9 | arXiv | 15/12/2025 | - | LADY: Linear Attention for Autonomous Driving Efficiency without Transformers |
| DiffRefiner | 90.7 | AAAI2026 | 17/11/2025 | coming soon | DiffRefiner: Coarse to Fine Trajectory Planning via Diffusion Refinement with Semantic Interaction for End to End Autonomous Driving |
| FutureX | 90.6 | arXiv | 11/12/2025 | coming soon | FutureX: Enhance End-to-End Autonomous Driving via Latent Chain-of-Thought World Model |
| UniDWM | 90.6 | arXiv | 01/02/2026 | 404 | UniDWM: Towards a Unified Driving World Model via Multifaceted Representation Learning |
| ResAD | 90.6 | arXiv | 08/10/2025 | coming soon | ResAD: Normalized Residual Trajectory Modeling for End-to-End Autonomous Driving |
| ExploreVLA | 90.4 | arXiv | 02/04/2026 | coming soon | ExploreVLA: Dense World Modeling and Exploration for End-to-End Autonomous Driving |
| EponaV2 | 36.1 | arXiv | - | coming soon | EponaV2: Driving World Model with Comprehensive Future Reasoning |
| SpanVLA | 90.3 | arXiv | 19/04/2026 | coming soon | SpanVLA: Efficient Action Bridging and Learning from Negative-Recovery Samples for Vision-Language-Action Model |
| AdaThinkDrive | 90.3 | CVPR2025 | 13/09/2025 | - | AdaThinkDrive: Adaptive Thinking via Reinforcement Learning for Autonomous Driving |
| HAD | 90.2 | arXiv | 03/04/2026 | - | HAD: Combining Hierarchical Diffusion with Metric-Decoupled RL for End-to-End Driving |
| FeaXDrive | 90.0 | NeurIPS2025 | 12/04/2026 | - | FeaXDrive: Feasibility-aware Trajectory-Centric Diffusion Planning for End-to-End Autonomous Driving |
| DriveDPO | 90.0 | NeurIPS2025 | 17/09/2025 | - | DriveDPO: Policy Learning via Safety DPO For End-to-End Autonomous Driving |
| - | 89.8 | arXiv | 20/09/2025 | - | Autoregressive End-to-End Planning with Time-Invariant Spatial Alignment and Multi-Objective Policy Refinement |
| CdDrive | 89.2 | arXiv | 03/02/2026 | 404 | A Unified Candidate Set with Scene-Adaptive Refinement via Diffusion for End-to-End Autonomous Driving |
| MindDrive | 88.9 | arXiv | 04/12/2025 | - | MindDrive: An All-in-One Framework Bridging World Models and Vision-Language Model for End-to-End Autonomous Driving |
| GMF-Drive | 88.9 | arXiv | 06/08/2025 | - | GMF-Drive: Gated Mamba Fusion with Spatial-Aware BEV Representation for End-to-End Autonomous Driving |
| Map-World | 88.8 | arXiv | 20/11/2025 | - | Map-World: Masked Action planning and Path-Integral World Model for Autonomous Driving |
| READ | 88.8 | - | 13/09/2025 | - | READ: End-to-End Autonomous Driving Made Safer with Efficient Reinforcement Learning |
| DynFlowDrive | 88.7 | - | 19/03/2026 | coming soon | DynFlowDrive: Flow-Based Dynamic World Modeling for Autonomous Driving |
| DreamerAD | 88.7 | - | 24/03/2026 | - | DreamerAD: Efficient Reinforcement Learning via Latent World Model for Autonomous Driving |
| TrajDiff | 88.5 | arXiv | 01/12/2025 | coming soon | TrajDiff: End-to-end Autonomous Driving without Perception Annotation |
| WorldRFT | 87.8 | AAAI2026 | 19/12/2025 | coming soon | WorldRFT: Latent World Model Planning with Reinforcement Fine-Tuning for Autonomous Driving |
| SUPER-AD | 87.7 | arXiv | 22/11/2025 | - | SUPER-AD: Semantic Uncertainty-aware Planning for End-to-End Robust Autonomous Driving |
| TrajHF | 87.6 | arXiv | 10/03/2025 | - | Learning Personalized Driving Styles via Reinforcement Learning from Human Feedback |
| CausalVAD | 87.2 | arXiv | 18/03/2026 | - | CausalVAD: De-confounding End-to-End Autonomous Driving via Causal Intervention |
| DAP | 87.2 | arXiv | 13/11/2025 | - | DAP: A Discrete-token Autoregressive Planner for Autonomous Driving |
| ARTEMIS | 87.0 | arXiv | 19/04/2025 | coming soon | ARTEMIS: Autoregressive End-to-End Trajectory Planning with Mixture of Experts for Autonomous Driving |
| ProDrive | 86.6 | arXiv | 25/04/2026 | - | ProDrive: Proactive Planning for Autonomous Driving via Ego-Environment Co-Evolution |
| NoRD | 85.6 | arXiv | 21/02/2026 | - | NoRD: A Data-Efficient Vision-Language-Action Model that Drives without Reasoning |
| LFG | 85.2 | arXiv | 22/02/2026 | - | Learning to Drive is a Free Gift: Large-Scale Label-Free Autonomy Pretraining from Unposed In-The-Wild Videos |
| DriveX | 84.5 | ICCV2025 | 19/05/2025 | - | DriveX: Omni Scene Modeling for Learning Generalizable World Knowledge in Autonomous Driving |
| DISK | 83.6 | arXiv | 01/02/2026 | - | DISK: Dynamic Inference SKipping for World Models |
| DrivingGPT | 82.4 | ICCV2025 | 18/12/2024 | 404 | DrivingGPT: Unifying Driving World Modeling and Planning with Multi-modal Autoregressive Transformers |