mlsanityData to 14 Oct 2026

DiDrive: A Risk-Aware Hierarchical Diffusion Framework for Safe Offline Reinforcement Learning in Autonomous Driving

arXiv 2609.01609 ↗First posted 4 Jun 2026cs.LG cs.ROcited by 0

On the map: Diffusion models → Diffusion and flow policies for robotics

also in 2 more
Abstract (arXiv)

While diffusion models effectively capture multimodal behavioral priors for autonomous driving, offline reinforcement learning (RL) policies remain susceptible to distribution shift, heavy-tailed risk signals, out-of-distribution (OOD) action generation, and high-dimensional state redundancy. To address these challenges, we propose DiDrive, a distribution-guided offline diffusion framework featuring two synergistic components: the Risk-Aware Hierarchical Diffusion (RHDif) architecture and the 3DICE policy optimization paradigm. In the state space, RHDif utilizes a low-level risk-gated encoder and a high-level contextual modulator to filter environmental redundancy and focus on safety-critical threats. In the action space, 3DICE mitigates OOD overestimation and gradient oscillation through in-sample calibrated guidance, spatiotemporal optimization, and ensemble-based candidate ranking. Evaluations on the CARLA benchmark demonstrate DiDrive's superiority over baselines like IQL, CQL, and Diffusion-QL, particularly in complex, high-density traffic scenarios with 60 vehicles, where it achieves an 85% success rate and a 4295.68 average reward, providing a robust pathway for safe autonomous driving decision-making.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Placed in Diffusion and flow policies for robotics, one of 14 directions in Diffusion models. Builds on 42 of its 46 references that are in the map. First posted on arXiv 4 Jun 2026.

Builds on

42 of 46 references.

Closest ancestors · 6

Same direction, newest first

  1. 2025GoalFlow: Goal-Driven Flow Matching for Multimodal Trajectories Generation in End-to-End Autonomous Driving7 Mar
  2. 2024DiffusionDrive: Truncated Diffusion Model for End-to-End Autonomous Driving22 Nov
  3. 2023Efficient Diffusion Policies for Offline Reinforcement Learning31 May
  4. 2023Contrastive Energy Prediction for Exact Energy-Guided Diffusion Sampling in Offline Reinforcement Learning25 Apr
  5. 2023Diffusion Policy: Visuomotor Policy Learning via Action Diffusion7 Mar
  6. 2022Diffusion Policies as an Expressive Policy Class for Offline Reinforcement Learning12 Aug

Other references · 36

  1. 2020Denoising Diffusion Probabilistic Modelscited by 21,076
  2. 2020Score-Based Generative Modeling through Stochastic Differential Equationscited by 9,122
  3. 2021Diffusion Models Beat GANs on Image Synthesiscited by 8,255
  4. 2019nuScenes: A multimodal dataset for autonomous drivingcited by 4,523
  5. 2017CARLA: An Open Urban Driving Simulatorcited by 2,449
  6. 2020Conservative Q-Learning for Offline Reinforcement Learningcited by 1,809
  7. 2021Decision Transformer: Reinforcement Learning via Sequence Modelingcited by 1,546
  8. 2021Offline Reinforcement Learning with Implicit Q-Learningcited by 1,293
  9. 2018Off-Policy Deep Reinforcement Learning without Explorationcited by 1,241
  10. 2022Planning-oriented Autonomous Drivingcited by 1,087
  11. 2021A Minimalist Approach to Offline Reinforcement Learningcited by 899
  12. 2019Stabilizing Off-Policy Q-Learning via Bootstrapping Error Reductioncited by 840
24 more
  1. 2023VAD: Vectorized Scene Representation for Efficient Autonomous Drivingcited by 622
  2. 2021Large Scale Interactive Motion Forecasting for Autonomous Driving : The Waymo Open Motion Datasetcited by 537
  3. 2019DualDICE: Behavior-Agnostic Estimation of Discounted Stationary Distribution Correctionscited by 254
  4. 2024Bench2Drive: Towards Multi-Ability Benchmarking of Closed-Loop End-To-End Autonomous Drivingcited by 252
  5. 2023Is Ego Status All You Need for Open-Loop End-to-End Autonomous Driving?cited by 242
  6. 2022A Survey on Offline Reinforcement Learning: Taxonomy, Review, and Open Problemscited by 231
  7. 2024GenAD: Generative End-to-End Autonomous Drivingcited by 225
  8. 2023MotionDiffuser: Controllable Multi-Agent Motion Prediction using Diffusioncited by 213
  9. 2024DriveDreamer-2: LLM-Enhanced World Models for Diverse Driving Video Generationcited by 191
  10. 2023Think Twice before Driving: Towards Scalable Decoders for End-to-End Autonomous Drivingcited by 169
  11. 2023A Survey of Imitation Learning: Algorithms, Recent Developments, and Challengescited by 148
  12. 2025Diffusion-Based Planning for Autonomous Driving with Flexible Guidancecited by 141
  13. 2020GenDICE: Generalized Offline Estimation of Stationary Valuescited by 122
  14. 2021OptiDICE: Offline Policy Optimization via Stationary Distribution Correction Estimationcited by 121
  15. 2022Supported Policy Optimization for Offline Reinforcement Learningcited by 74
  16. 2023Elastic Decision Transformercited by 67
  17. 2024DiffuserLite: Towards Real-time Diffusion Planningcited by 42
  18. 2023Policy Regularization with Dataset Constraint for Offline Reinforcement Learningcited by 34
  19. 2025SceneDiffuser++: City-Scale Traffic Simulation via a Generative World Modelcited by 20
  20. 2024Offline Reinforcement Learning with OOD State Correction and OOD Action Suppressioncited by 17
  21. 2021DQ-GAT: Towards Safe and Efficient Autonomous Driving with Deep Q-Learning and Graph Attention Networkscited by 13
  22. 2024RACP: Risk-Aware Contingency Planning with Multi-Modal Predictionscited by 10
  23. 2023Robust Offline Reinforcement learning with Heavy-Tailed Rewardscited by 7
  24. 2025From Imitation to Optimization: A Comparative Study of Offline Learning for Autonomous Drivingcited by 4

Built on by

Not cited yet.

Similar, not cited

Close in wording, no citation either way

  1. 2025DiffusionDriveV2: Reinforcement Learning-Constrained Truncated Diffusion Modeling in End-to-End Autonomous Driving
  2. 2026HAD: Combining Hierarchical Diffusion with Metric-Decoupled RL for End-to-End Driving
  3. 2023Safety-aware Causal Representation for Trustworthy Offline Reinforcement Learning in Autonomous Driving
  4. 2021UMBRELLA: Uncertainty-Aware Model-Based Offline Reinforcement Learning Leveraging Planning
  5. 2025AD-R1: Closed-Loop Reinforcement Learning for End-to-End Autonomous Driving with Impartial World Models
  6. 2023Guided Online Distillation: Promoting Safe Reinforcement Learning by Offline Demonstration
  7. 2026Unleashing the Potential of Diffusion Models for End-to-End Autonomous Driving
  8. 2026BridgeGuard: Explicit Safety Drift for Diffusion-based Autonomous Driving