I am currently a first-year PhD student at the National University of Singapore, supervised by Prof. Mengling Feng.
Reviewer for NIPS, IJCAI, ACM TIST, with over 10 papers reviewed.
Research Interests: Trustworthy and Agentic LLM • Multi-modality Intelligence in Healthcare
Looking for a summer 26' internship!! 🙏
我目前是新加坡国立大学一年级博士生，导师为 冯梦凌教授。
担任NIPS、IJCAI、ACM TIST等会议和期刊的审稿人，已审稿10余篇。
研究方向：可信与智能体大语言模型 • 医疗领域多模态智能
期待26年暑期实习机会！！🙏
Warning
Problem: The current name of your GitHub Pages repository ("Solution: Please consider renaming the repository to "
http://".
However, if the current repository name is intended, you can ignore this message by removing "{% include widgets/debug_repo_name.html %}" in index.html.
Action required
Problem: The current root path of this site is "baseurl ("_config.yml.
Solution: Please set the
baseurl in _config.yml to "Education 教育经历
-
National University of SingaporePh.D. in Artificial IntelligenceJan. 2025 - present -
The University of Hong KongM.Sc. in Artificial IntelligenceSep. 2022 - Jul. 2024 -
The Chinese University of Hong Kong, ShenzhenB.Sc. in Statistics, Data Science StreamSep. 2018 - May. 2022
-
新加坡国立大学人工智能博士2025年1月 - 至今
-
香港大学人工智能理学硕士2022年9月 - 2024年7月
-
香港中文大学（深圳）统计学理学学士（数据科学方向）2018年9月 - 2022年5月
Experience 经历
-
StepFun AI Intelligent TechnologyLLM Research InternJan. 2025 - June. 2025 -
The Chinese University of Hong KongResearch StaffSep. 2024 - Jan. 2025
-
阶跃星辰（StepFun）大模型实习生2025年1月 - 2025年6月
-
香港中文大学科研助理2024年9月 - 2025年1月
Honors & Awards 荣誉与奖励
-
Full Ph.D. Scholarship, National University of Singapore2025
-
Outstanding College Graduate, CUHK-Shenzhen Harmonia College2022
-
Undergraduate Research Excellence Award, CUHK-Shenzhen2021
-
新加坡国立大学全额博士奖学金2025
-
香港中文大学（深圳）祥波书院优秀毕业生2022
-
香港中文大学（深圳）本科生科研卓越奖2021
Selected Publications (view all ) （查看全部 ）
DivScore: Zero-Shot Detection of LLM-Generated Text in Specialized Domains 🔗
Zhihui Chen, Kai He, Yucheng Huang, Yunxiao Zhu, Mengling Feng
Conference on Empirical Methods in Natural Language Processing (EMNLP) 2025
Detecting LLM-generated text in specialized and high-stakes domains like medicine and law is crucial for combating misinformation and ensuring authenticity. We propose DivScore, a zero-shot detection framework using normalized entropy-based scoring and domain knowledge distillation to robustly identify LLM-generated text in specialized domains. Experiments show that DivScore consistently outperforms state-of-the-art detectors, with 14.4% higher AUROC and 64.0% higher recall at 0.1% false positive rate threshold.
DivScore: Zero-Shot Detection of LLM-Generated Text in Specialized Domains 🔗
Zhihui Chen, Kai He, Yucheng Huang, Yunxiao Zhu, Mengling Feng
Conference on Empirical Methods in Natural Language Processing (EMNLP) 2025
Detecting LLM-generated text in specialized and high-stakes domains like medicine and law is crucial for combating misinformation and ensuring authenticity. We propose DivScore, a zero-shot detection framework using normalized entropy-based scoring and domain knowledge distillation to robustly identify LLM-generated text in specialized domains. Experiments show that DivScore consistently outperforms state-of-the-art detectors, with 14.4% higher AUROC and 64.0% higher recall at 0.1% false positive rate threshold.
Med-Banana-50K: A Large-Scale Cross-Modality Dataset for Medical Image Editing 🔗
Zhihui Chen, et al.
arXiv preprint 2025
Recent advances in multimodal large language models have enabled remarkable medical image editing capabilities. However, the research community's progress remains constrained by the absence of large-scale, high-quality, and openly accessible datasets built specifically for medical image editing with strict anatomical and clinical constraints. We introduce Med-Banana-50K, a comprehensive 50K-image dataset for instruction-based medical image editing spanning three modalities (chest X-ray, brain MRI, fundus photography) and 23 disease types. Our dataset is constructed by leveraging Gemini-2.5-Flash-Image to generate bidirectional edits (lesion addition and removal) from real medical images. What distinguishes Med-Banana-50K from general-domain editing datasets is our systematic approach to medical quality control: we employ LLM-as-Judge with a medically grounded rubric and history-aware iterative refinement up to five rounds.
Med-Banana-50K: A Large-Scale Cross-Modality Dataset for Medical Image Editing 🔗
Zhihui Chen, et al.
arXiv preprint 2025
Recent advances in multimodal large language models have enabled remarkable medical image editing capabilities. However, the research community's progress remains constrained by the absence of large-scale, high-quality, and openly accessible datasets built specifically for medical image editing with strict anatomical and clinical constraints. We introduce Med-Banana-50K, a comprehensive 50K-image dataset for instruction-based medical image editing spanning three modalities (chest X-ray, brain MRI, fundus photography) and 23 disease types. Our dataset is constructed by leveraging Gemini-2.5-Flash-Image to generate bidirectional edits (lesion addition and removal) from real medical images. What distinguishes Med-Banana-50K from general-domain editing datasets is our systematic approach to medical quality control: we employ LLM-as-Judge with a medically grounded rubric and history-aware iterative refinement up to five rounds.