ForSim: Stepwise Forward Simulation for Traffic Policy Fine-Tuning

Keyu Chen, Wenchao Sun, Hao Cheng, Zheng Fu, Sifa Zheng

ICRA 2026

ForSim achieves high-fidelity traffic simulation by combining stepwise closed-loop trajectory propagation with physical dynamics for multimodal diversity and interaction-aware agent updates for realistic behavioral consistency.

RIFT: Group-Relative RL Fine-Tuning for Realistic and Controllable Traffic Simulation

Keyu Chen, Wenchao Sun, Hao Cheng, Sifa Zheng

arXiv 2025

RIFT achieve realistic and controllable traffic simulation by combining IL pre-training in a data-driven simulator for realism with RL fine-tuning in a physics-based simulator for controllability.

FREA: Feasibility-Guided Generation of Safety-Critical Scenarios with Reasonable Adversariality

Keyu Chen, Yuheng Lei, Hao Cheng, Haoran Wu, Wenchao Sun, Sifa Zheng

Conference on Robot Learning (CoRL Oral 4.3%), 2024

FREA incorporates feasibility as guidance to generate adversarial yet AV-feasible, safety-critical scenarios for autonomous driving.

IGT: Illumination-guided RGB-T object detection with transformers

Keyu Chen, Jinqiang Liu, Han Zhang

Knowledge-Based Systems (KBS) (SCI Q1, IF=7.4), 2023

IGT uses illumination intensity to guide the fusion process of multi-modality features, enabling the comprehensive utilization of cross-modality complementary information in object detection.

Highway-Adversarial

We manipulate critical background vehicles in the Highway-Env to exhibit adversarial behaviors that challenge the ego vehicle, creating safety-critical scenarios.