Achieving both realism and controllability in closed-loop traffic simulation remains a key challenge in autonomous driving. Dataset-based methods reproduce realistic trajectories but suffer from covariate shift in closed-loop deployment, compounded by simplified dynamics models that further reduce reliability. Conversely, physics-based simulation methods enhance reliable and controllable closed-loop interactions but often lack expert demonstrations, compromising realism. To address these challenges, we introduce a dual-stage AV-centric simulation framework that conducts imitation learning pre-training in a data-driven simulator to capture trajectory-level realism and route-level controllability, followed by reinforcement learning fine-tuning in a physics-based simulator to enhance style-level controllability and mitigate covariate shift. In the fine-tuning stage, we propose RIFT, a novel group-relative RL fine-tuning strategy that evaluates all candidate modalities through group-relative formulation and employs a surrogate objective for stable optimization, enhancing style-level controllability and mitigating covariate shift while preserving the trajectory-level realism and route-level controllability inherited from IL pre-training. Extensive experiments demonstrate that RIFT improves realism and controllability in traffic simulation while simultaneously exposing the limitations of modern AV systems in closed-loop evaluation. Project Page: https://currychen77.github.io/RIFT/
Controllable Traffic Simulation through LLM-Guided Hierarchical Reasoning and Refinement
School of Vehicle and Mobility, Tsinghua University(清华大学车辆与移动学院);Artificial Intelligence Thrust, The Hong Kong University of Science and Technology (Guangzhou)(香港理工大学(广州)人工智能研究部);School of Computer Science, Carnegie Mellon University(卡内基梅隆大学计算机科学学院);Department of Computer Sciences, University of Wisconsin-Madison(威斯康星大学麦迪逊分校计算机科学系)
LCSim: A Large-Scale Controllable Traffic Simulator
Tsinghua University(清华大学);Beijing National Research Center for Information Science and Technology (BNRist)(北京信息科学与技术国家研究中心);Shanghai Lingang Senseauto Intelligent Technology Co., Ltd.(上海临港智己智能科技有限公司)
HAD-Gen: Human-like and Diverse Driving Behavior Modeling for Controllable Scenario Generation
School of Engineering and Physical Sciences, Heriot-Watt University(赫瑞斯泰德大学工程与物理科学学院);School of Automation and Software Engineering, Shanxi University(山西大学自动化与软件工程学院);School of Informatics, University of Edinburgh(爱丁堡大学信息学院)
DragTraffic: Interactive and Controllable Traffic Scene Generation for Autonomous Driving
The Hong Kong University of Science and Technology(香港科技大学);The Hong Kong University of Science and Technology (Guangzhou)(香港科技大学(广州));KTH Royal Institute of Technology(瑞典皇家理工学院);Lotus Technology Ltd(路特斯科技有限公司)
Versatile Behavior Diffusion for Generalized Traffic Agent Simulation
School of Mechanical and Aerospace Engineering, Nanyang Technological University(南洋理工大学机械与航空航天工程学院);Department of Electrical and Computer Engineering, Princeton University(普林斯顿大学电气与计算机工程系);NVIDIA Research(NVIDIA研究)
SymDrive: Realistic and Controllable Driving Simulator via Symmetric Auto-regressive Online Restoration
School of Vehicle and Mobility, Tsinghua University(清华大学车辆与移动系统学院);Shanghai Artificial Intelligence Laboratory(上海人工智能实验室);State Key Lab of Intelligent Transportation System, Beihang University(北航智能交通系统国家重点实验室)
Safety-Critical Traffic Simulation with Guided Latent Diffusion Model
The Hong Kong University of Science and Technology (Guangzhou)(香港科技大学(广州));School of Computer Science and Engineering(计算机科学与工程学院);Sun Yat-sen University(中山大学)
RADE: Learning Risk-Adjustable Driving Environment via Multi-Agent Conditional Diffusion
Department of Civil and Environmental Engineering, University of Michigan(密歇根大学土木与环境工程系);Department of Computer Science, University of Hong Kong(香港大学计算机科学系);University of Michigan Transportation Research Institute(密歇根大学交通研究研究院)