people
members of the lab or group
555 your office number
123 your address street
Your City, State 12345
I am a 2nd-year Ph.D. student in Computer Science at the University of Maryland (UMD), advised by Prof. Soheil Feizi. My research focuses on the safety and reliability of LLMs and LLM agents, particularly in evaluation, reasoning, controllability, and deployment.
- Reliable tool use in LLM agents: studying unstable tool preferences (EMNLP’25), temporally misaligned tool-use decisions (ACL’26), and the gap between recognizing and acting on tool necessity (arXiv’26).
- Reliable evaluation and control: test-set contamination detection (EMNLP’25), red-teaming AI-text detection (NeurIPS’25), and inference-time activation steering (arXiv’26).
- LLM reasoning: interpretable mathematical reasoning (ACL’26) and exploration in interactive tasks (arXiv’26).
Prior to joining UMD, I earned my bachelor’s degree in Computer Science and Electrical Engineering from the Hong Kong University of Science and Technology (HKUST). During my undergraduate studies, I had the opportunity to work with Prof. Minhao Cheng on backdoor attacks and machine learning watermarks. I also spent a semester on exchange at ETH Zurich, where I was fortunate to have worked with Prof. Florian Tramèr on adversarial examples and diffusion models.
555 your office number
123 your address street
Your City, State 12345
I am a 2nd-year Ph.D. student in Computer Science at the University of Maryland (UMD), advised by Prof. Soheil Feizi. My research focuses on the safety and reliability of LLMs and LLM agents, particularly in evaluation, reasoning, controllability, and deployment.
- Reliable tool use in LLM agents: studying unstable tool preferences (EMNLP’25), temporally misaligned tool-use decisions (ACL’26), and the gap between recognizing and acting on tool necessity (arXiv’26).
- Reliable evaluation and control: test-set contamination detection (EMNLP’25), red-teaming AI-text detection (NeurIPS’25), and inference-time activation steering (arXiv’26).
- LLM reasoning: interpretable mathematical reasoning (ACL’26) and exploration in interactive tasks (arXiv’26).
Prior to joining UMD, I earned my bachelor’s degree in Computer Science and Electrical Engineering from the Hong Kong University of Science and Technology (HKUST). During my undergraduate studies, I had the opportunity to work with Prof. Minhao Cheng on backdoor attacks and machine learning watermarks. I also spent a semester on exchange at ETH Zurich, where I was fortunate to have worked with Prof. Florian Tramèr on adversarial examples and diffusion models.