About Me

Hi👋I’m Zhaohan, third-year PhD student from EECS Department in Queen Mary University of London (QMUL). I am fortunate to be supervised by Prof. Ioannis Patras and Dr. Ziquan Liu.

Before beginning my Ph.D. journey, I obtained my Bachelor’s and Master’s degrees from Xi’an Jiaotong University, where I first became interested in AI and large language models under the insightful and generous guidance of Prof. Xiaoming Liu, Prof. Chao Shen, and Prof. Xiaohong Guan.

Research Interest

My research lies at the intersection of natural language processing and machine learning, with a focus on the reliability, safety, and efficiency in the deployment of Large Language Models (LLMs). I am currently exploring:

• Confidence elicitation, calibration, and its application in LLMs.
• Effective and robust AI-Generated Content (AIGC) detection methods.
• Efficient reasoning scheme.

News

Preprints

• GrACE: A Generative Approach to Better Confidence Elicitation in Large Language Models
  Zhaohan Zhang, Ziquan Liu, Ioannis Patras
  

• Upfront Chain-of-Thought: A Cooperative Framework for Chain-of-Thought Compression
  Chengzhengxu Li, Xiaoming Liu, Zhaohan Zhang, Shaochu Zhang, Shengchao Liu, Guoxin Ma, Yu Lan, Chao Shen
  

• DEER: Disentangled Mixture of Experts with Instance-Adaptive Routing for Generalizable Machine-Generated Text Detection
  Guoxin Ma, Xiaoming Liu, Zhanhan Zhang, Chengzhengxu Li, Shengchao Liu, Yu Lan
  

Selected Publications

• Get confused cautiously: Textual sequence memorization erasure with selective entropy maximization
  Zhaohan Zhang, Ziquan Liu, Ioannis Patras
  in COLING2025 (Oral Presentation).

• Iron Sharpens Iron: Defending Against Attacks in Machine-Generated Text Detection with Adversarial Training
  Yuanfan Li*, Zhaohan Zhang*, Chengzhengxu Li*, Chao Shen, Xiaoming Liu
  in ACL2025.

• Coco: Coherence-enhanced machine-generated text detection under low resource with contrastive learning
  Xiaoming Liu*, Zhaohan Zhang*, Yichen Wang*, Hang Pu, Yu Lan, Chao Shen
  in EMNLP2023.

• MGT-Prism: Enhancing Domain Generalization for Machine-Generated Text Detection via Spectral Alignment
  Shengchao Liu, Xiaoming Liu, Chengzhengxu Li, Zhaohan Zhang, Guoxin Ma, Yu Lan, Shuai Xiao
  in AAAI2026.

• Concentrate Attention: Towards Domain-Generalizable Prompt Optimization for Language Models
  Chengzhengxu Li, Xiaoming Liu, Zhaohan Zhang, Yichen Wang, Chen Liu, Yu Lan, Chao Shen
  in NeurIPS2024.

• Stablept: Towards stable prompting for few-shot learning via input separation
  Xiaoming Liu, Chen Liu, Zhaohan Zhang, Chengzhengxu Li, Longtian Wang, Yu Lan, Chao Shen
  in Findings of EMNLP2024.

Educations

• Sep 2023 – present · Queen Mary University of London🇬🇧 — Ph.D. Student
• Aug 2020 – Jun 2023 · Xi’an Jiaotong University🇨🇳 — M.Eng. (Graduate with Honor)
• Aug 2018 – May 2019 · University of Wisconsin–Madison🇺🇸 — Visiting Student
• Aug 2016 – Jun 2020 · Xi’an Jiaotong University🇨🇳 — B.Eng. (Graduate with Honor)

Invited Talks

• 2025.11. GrACE: A Generative Approach to Better Confidence Elicitation in Large Language Models, in BMVA Symposium on Multimodal Large Models

Services

Reviewer: ACL Rolling Review (2024-), NeurIPS (2023-), ICML (2024-), ICLR (2025-)