{"slug": "didrive-a-risk-aware-hierarchical-diffusion-framework-for-safe-offline-learning", "title": "DiDrive: A Risk-Aware Hierarchical Diffusion Framework for Safe Offline Reinforcement Learning in Autonomous Driving", "summary": "Researchers propose DiDrive, a risk-aware hierarchical diffusion framework for safe offline reinforcement learning in autonomous driving, which achieves an 85% success rate and a 4295.68 average reward in CARLA benchmark simulations with 60 vehicles, outperforming baselines like IQL, CQL, and Diffusion-QL. The framework combines the Risk-Aware Hierarchical Diffusion (RHDif) architecture and the 3DICE policy optimization paradigm to address distribution shift, OOD action generation, and high-dimensional state redundancy.", "body_md": "arXiv:2609.01609v1 Announce Type: new\nAbstract: 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.", "url": "https://wpnews.pro/news/didrive-a-risk-aware-hierarchical-diffusion-framework-for-safe-offline-learning", "canonical_source": "https://arxiv.org/abs/2609.01609", "published_at": "2026-09-03 04:00:00+00:00", "updated_at": "2026-09-03 04:23:06.019896+00:00", "lang": "en", "topics": ["artificial-intelligence", "machine-learning", "autonomous-vehicles", "ai-research"], "entities": ["DiDrive", "CARLA", "IQL", "CQL", "Diffusion-QL", "RHDif", "3DICE"], "alternates": {"html": "https://wpnews.pro/news/didrive-a-risk-aware-hierarchical-diffusion-framework-for-safe-offline-learning", "markdown": "https://wpnews.pro/news/didrive-a-risk-aware-hierarchical-diffusion-framework-for-safe-offline-learning.md", "text": "https://wpnews.pro/news/didrive-a-risk-aware-hierarchical-diffusion-framework-for-safe-offline-learning.txt", "jsonld": "https://wpnews.pro/news/didrive-a-risk-aware-hierarchical-diffusion-framework-for-safe-offline-learning.jsonld"}}