# DriveDPO: Policy Learning via Safety DPO For End-to-End Autonomous Driving

arXiv: https://arxiv.org/abs/2509.17940 · HTML: https://mlsanity.com/paper/2509.17940/

## Key facts

- Cited by 39 papers in the map (data through 14 Oct 2026).
- Placed in Imitation-learning end-to-end planners, one of 9 directions in End-to-end autonomous driving.
- Builds on 51 of its 59 references that are in the map.
- First posted on arXiv 22 Sep 2025.

## Placed in

- End-to-end autonomous driving / [Imitation-learning end-to-end planners](https://mlsanity.com/driving/imitation-learning-end-to-end-planners/)
- LLM alignment / RLHF / preference optimisation / [Reward modeling and RLHF](https://mlsanity.com/alignment/reward-modeling-and-rlhf/)

## Builds on (most cited first, 30 of 51)

- [Attention Is All You Need](https://mlsanity.com/paper/1706.03762/) (2017)
- [Deep Residual Learning for Image Recognition](https://mlsanity.com/paper/1512.03385/) (2015)
- [GPT-4 Technical Report](https://mlsanity.com/paper/2303.08774/) (2023)
- [Training language models to follow instructions with human feedback](https://mlsanity.com/paper/2203.02155/) (2022)
- [Direct Preference Optimization: Your Language Model is Secretly a Reward Model](https://mlsanity.com/paper/2305.18290/) (2023)
- [Deep reinforcement learning from human preferences](https://mlsanity.com/paper/1706.03741/) (2017)
- [Gradient Surgery for Multi-Task Learning](https://mlsanity.com/paper/2001.06782/) (2020)
- [Planning-oriented Autonomous Driving](https://mlsanity.com/paper/2212.10156/) (2022)
- [SimPO: Simple Preference Optimization with a Reference-Free Reward](https://mlsanity.com/paper/2405.14734/) (2024)
- [A General Theoretical Paradigm to Understand Learning from Human Preferences](https://mlsanity.com/paper/2310.12036/) (2023)
- [Multi-Task Learning as Multi-Objective Optimization](https://mlsanity.com/paper/1810.04650/) (2018)
- [RLAIF vs. RLHF: Scaling Reinforcement Learning from Human Feedback with AI Feedback](https://mlsanity.com/paper/2309.00267/) (2023)
- [VAD: Vectorized Scene Representation for Efficient Autonomous Driving](https://mlsanity.com/paper/2303.12077/) (2023)
- [OpenScene: 3D Scene Understanding with Open Vocabularies](https://mlsanity.com/paper/2211.15654/) (2022)
- [TransFuser: Imitation with Transformer-Based Sensor Fusion for Autonomous Driving](https://mlsanity.com/paper/2205.15997/) (2022)
- [Contrastive Preference Optimization: Pushing the Boundaries of LLM Performance in Machine Translation](https://mlsanity.com/paper/2401.08417/) (2024)
- [ST-P3: End-to-end Vision-based Autonomous Driving via Spatial-Temporal Feature Learning](https://mlsanity.com/paper/2207.07601/) (2022)
- [NAVSIM: Data-Driven Non-Reactive Autonomous Vehicle Simulation and Benchmarking](https://mlsanity.com/paper/2406.15349/) (2024)
- [Iterative Preference Learning from Human Feedback: Bridging Theory and Practice for RLHF under KL-Constraint](https://mlsanity.com/paper/2312.11456/) (2023)
- [DiffusionDrive: Truncated Diffusion Model for End-to-End Autonomous Driving](https://mlsanity.com/paper/2411.15139/) (2024)
- [VADv2: End-to-End Vectorized Autonomous Driving via Probabilistic Planning](https://mlsanity.com/paper/2402.13243/) (2024)
- [Learning to Drive in a Day](https://mlsanity.com/paper/1807.00412/) (2018)
- [Is DPO Superior to PPO for LLM Alignment? A Comprehensive Study](https://mlsanity.com/paper/2404.10719/) (2024)
- [Bench2Drive: Towards Multi-Ability Benchmarking of Closed-Loop End-To-End Autonomous Driving](https://mlsanity.com/paper/2406.03877/) (2024)
- [Hydra-MDP: End-to-end Multimodal Planning with Multi-target Hydra-Distillation](https://mlsanity.com/paper/2406.06978/) (2024)
- [Parting with Misconceptions about Learning-based Vehicle Motion Planning](https://mlsanity.com/paper/2306.07962/) (2023)
- [SparseDrive: End-to-End Autonomous Driving via Sparse Scene Representation](https://mlsanity.com/paper/2405.19620/) (2024)
- [ReCogDrive: A Reinforced Cognitive Framework for End-to-End Autonomous Driving](https://mlsanity.com/paper/2506.08052/) (2025)
- [Enhancing End-to-End Autonomous Driving with Latent World Model](https://mlsanity.com/paper/2406.08481/) (2024)
- [Perceive, Predict, and Plan: Safe Motion Planning Through Interpretable Semantic Representations](https://mlsanity.com/paper/2008.05930/) (2020)

## Built on by (most cited first, 30 of 39)

- [Drive-JEPA: Video JEPA Meets Multimodal Trajectory Distillation for End-to-End Driving](https://mlsanity.com/paper/2601.22032/) (2026)
- [SpaceDrive: Infusing Spatial Awareness into VLM-based Autonomous Driving](https://mlsanity.com/paper/2512.10719/) (2025)
- [MindDrive: A Vision-Language-Action Model for Autonomous Driving via Online Reinforcement Learning](https://mlsanity.com/paper/2512.13636/) (2025)
- [UniDriveVLA: Unifying Understanding, Perception, and Action Planning for Autonomous Driving](https://mlsanity.com/paper/2604.02190/) (2026)
- [AD-R1: Closed-Loop Reinforcement Learning for End-to-End Autonomous Driving with Impartial World Models](https://mlsanity.com/paper/2511.20325/) (2025)
- [DynVLA: Learning World Dynamics for Action Reasoning in Autonomous Driving](https://mlsanity.com/paper/2603.11041/) (2026)
- [SafeDrive: Fine-Grained Safety Reasoning for End-to-End Driving in a Sparse World](https://mlsanity.com/paper/2602.18887/) (2026)
- [RAD-2: Scaling Reinforcement Learning in a Generator-Discriminator Framework](https://mlsanity.com/paper/2604.15308/) (2026)
- [The DAWN of World-Action Interactive Models](https://mlsanity.com/paper/2605.11550/) (2026)
- [Beyond Imitation: Learning Safe End-to-End Autonomous Driving from Hard Negatives](https://mlsanity.com/paper/2605.19771/) (2026)
- [Do Open-Loop Metrics Predict Closed-Loop Driving? A Cross-Benchmark Correlation Study of NAVSIM and Bench2Drive](https://mlsanity.com/paper/2605.00066/) (2026)
- [GraphWorld: Long-Horizon Planning with World Models for End-to-End Autonomous Driving](https://mlsanity.com/paper/2606.16274/) (2026)
- [CausalDrive: Real-time Causal World Models for Autonomous Driving](https://mlsanity.com/paper/2606.15341/) (2026)
- [WPT: World-to-Policy Transfer via Online World Model Distillation](https://mlsanity.com/paper/2511.20095/) (2025)
- [Fine-tuning is Not Enough: A Parallel Framework for Collaborative Imitation and Reinforcement Learning in End-to-end Autonomous Driving](https://mlsanity.com/paper/2603.13842/) (2026)
- [RISE: Adaptive Imagination for World Action Models](https://mlsanity.com/paper/2608.20430/) (2026)
- [Distill to Think, Foresee to Act: Cognitive-Physical Reinforcement Learning for Autonomous Driving](https://mlsanity.com/paper/2605.21139/) (2026)
- [VLGA: Vision-Language-Geometry-Action Models for Autonomous Driving](https://mlsanity.com/paper/2606.12396/) (2026)
- [SafeAlign-VLA: A Negative-Enhanced Safe Alignment Framework for Risk-Aware Autonomous Driving](https://mlsanity.com/paper/2605.19524/) (2026)
- [Intend, Reflect, Refine: An Adaptive Multimodal Reflection Framework for Autonomous Driving](https://mlsanity.com/paper/2606.22913/) (2026)
- [Post-Training in End-to-End Autonomous Driving](https://mlsanity.com/paper/2607.08072/) (2026)
- [A Unified Candidate Set with Scene-Adaptive Refinement via Diffusion for End-to-End Autonomous Driving](https://mlsanity.com/paper/2602.03112/) (2026)
- [Risk-Aware World Model Predictive Control for Generalizable End-to-End Autonomous Driving](https://mlsanity.com/paper/2602.23259/) (2026)
- [Causality-Aware End-to-End Autonomous Driving via Ego-Centric Joint Scene Modeling](https://mlsanity.com/paper/2605.13646/) (2026)
- [DOG-DPO:Dynamic Optimization in Geometry for Safety Alignment](https://mlsanity.com/paper/2606.07678/) (2026)
- [UniTeD: Unified Temporal Diffusion for Joint Perception and Planning in Autonomous Driving](https://mlsanity.com/paper/2606.25736/) (2026)
- [Learning from Mistakes: Rollout-Retrieval Lifelong Policy Learning for Autonomous Driving](https://mlsanity.com/paper/2606.30537/) (2026)
- [Unified Prediction and Planning via Conflict-Aware Disjoint Parameter Training](https://mlsanity.com/paper/2607.19971/) (2026)
- [GSDrive: Reinforcing Driving Policies by Multi-mode Future Trajectory Probing with 3D Gaussian Splatting Environment](https://mlsanity.com/paper/2604.28111/) (2026)
- [DriveSafer: End-to-End Autonomous Driving with Safety Guidance](https://mlsanity.com/paper/2605.16737/) (2026)

## Abstract (arXiv)

End-to-end autonomous driving has substantially progressed by directly predicting future trajectories from raw perception inputs, which bypasses traditional modular pipelines. However, mainstream methods trained via imitation learning suffer from critical safety limitations, as they fail to distinguish between trajectories that appear human-like but are potentially unsafe. Some recent approaches attempt to address this by regressing multiple rule-driven scores but decoupling supervision from policy optimization, resulting in suboptimal performance. To tackle these challenges, we propose DriveDPO, a Safety Direct Preference Optimization Policy Learning framework. First, we distill a unified policy distribution from human imitation similarity and rule-based safety scores for direct policy optimization. Further, we introduce an iterative Direct Preference Optimization stage formulated as trajectory-level preference alignment. Extensive experiments on the NAVSIM benchmark demonstrate that DriveDPO achieves a new state-of-the-art PDMS of 90.0. Furthermore, qualitative results across diverse challenging scenarios highlight DriveDPO's ability to produce safer and more reliable driving behaviors.