4 papers
CRAFT: Counterfactual-to-Interactive Reinforcement Fine-Tuning for Driving Policies
Keyu Chen, Nanfei Ye, Yida Wang +4
Open-loop imitation learning has advanced modern autonomous driving policy architectures, but closed-loop deployment remains vulnerable to policy-induced distribution shift. Existi…
ForSim: Stepwise Forward Simulation for Traffic Policy Fine-Tuning
Keyu Chen, Wenchao Sun, Hao Cheng +2
As the foundation of closed-loop training and evaluation in autonomous driving, traffic simulation still faces two fundamental challenges: covariate shift introduced by open-loop i…
Modified-Emergency Index (MEI): A Criticality Metric for Autonomous Driving in Lateral Conflict
Hao Cheng, Yanbo Jiang, Qingyuan Shi +5
Effective, reliable, and efficient evaluation of autonomous driving safety is essential to demonstrate its trustworthiness. Criticality metrics provide an objective means of assess…
RIFT: Group-Relative RL Fine-Tuning for Realistic and Controllable Traffic Simulation
Keyu Chen, Wenchao Sun, Hao Cheng +1
Achieving both realism and controllability in closed-loop traffic simulation remains a key challenge in autonomous driving. Dataset-based methods reproduce realistic trajectories b…