Kun Zhan
Li Auto Inc.
収録論文 18本 ・ フィジカルAI/ロボット学習
自動運転
※arXiv著者名で収集。同姓同名の別人の論文が含まれる場合があります。
論文
- BrainWAM: 自動運転のための意味的先行知識と予測ダイナミクスの行動空間協調自動運転2026/8/13
自動運転の計画において、意味的推論（VLA）と予測的ダイナミクス（WAM）を統合する際の注意配分の不一致を解決するため、行動空間で協調させるBrainWAMを提案し、NAVSIMで最高性能を達成した。
- BrainWAM: Action-Space Coordination of Semantic Priors and Predictive Dynamics for Autonomous Driving2026/8/1
- VLAFlow: A Unified Training Framework for Vision-Language-Action Models via Co-training and Future Latent Alignment2026/7/1
- Closed Loop Dynamic Driving Data Mixture for Real-Synthetic Co-Training2026/5/1
- ReflectDrive-2: Reinforcement-Learning-Aligned Self-Editing for Discrete Diffusion Driving2026/5/1
- Unifying Language-Action Understanding and Generation for Autonomous Driving2026/3/1
- Hardware Co-Design Scaling Laws via Roofline Modelling for On-Device LLMs2026/2/10
- WorldRFT: Latent World Model Planning with Reinforcement Fine-Tuning for Autonomous Driving2025/12/1
- The Better You Learn, The Smarter You Prune: Towards Efficient Vision-language-action Models via Differentiable Token Pruning2025/9/1
- Discrete Diffusion for Reflective Vision-Language-Action Models in Autonomous Driving2025/9/1
- RoboPearls: Editable Video Simulation for Robot Manipulation2025/6/1
- GeoDrive: 3D Geometry-Informed Driving World Model with Precise Action Control2025/5/1
- TransDiffuser: Diverse Trajectory Generation with Decorrelated Multi-modal Representation for End-to-end Autonomous Driving2025/5/1
- Learning Personalized Driving Styles via Reinforcement Learning from Human Feedback2025/3/1
- ReconDreamer: Crafting World Models for Driving Scene Reconstruction via Online Restoration2024/11/1
- DrivingSphere: Building a High-fidelity 4D World for Closed-loop Simulation2024/11/1
- Generalizing Motion Planners with Mixture of Experts for Autonomous Driving2024/10/1
- PlanAgent: A Multi-modal Large Language Agent for Closed-loop Vehicle Motion Planning2024/6/1