5 papers
StratMem-Bench: Evaluating Strategic Memory Use in Virtual Character Conversation Beyond Factual Recall
Yerong Wu, Tianxing Wu, Minghao Zhu +2
Achieving realistic human-like conversation for virtual characters requires not only a simple memorization and recall of past events, but also the strategic utilization of memory t…
Beyond Semantic Manipulation: Token-Space Attacks on Reward Models
Yuheng Zhang, Mingyue Huo, Minghao Zhu +2
Reward models (RMs) are widely used as optimization targets in reinforcement learning from human feedback (RLHF), yet they remain vulnerable to reward hacking. Existing attacks mai…
SignRAG: A Retrieval-Augmented System for Scalable Zero-Shot Road Sign Recognition
Minghao Zhu, Zhihao Zhang, Anmol Sidhu +1
Automated road sign recognition is a critical task for intelligent transportation systems, but traditional deep learning methods struggle with the sheer number of sign classes and…
An Uncertainty-Weighted Decision Transformer for Navigation in Dense, Complex Driving Scenarios
Zhihao Zhang, Chengyang Peng, Minghao Zhu +2
Autonomous driving in dense, dynamic environments requires decision-making systems that can exploit both spatial structure and long-horizon temporal dependencies while remaining ro…
Training Adversarial yet Safe Agent to Characterize Safety Performance of Highly Automated Vehicles
Minghao Zhu, Anmol Sidhu, Keith A. Redmill
This paper focuses on safety performance testing and characterization of black-box highly automated vehicles (HAV). Existing testing approaches typically obtain the testing outcome…