Cheng Wang's Homepage

Cheng Wang

Hi, I am Cheng Wang (王程). I'm a final-year undergraduate student from National University of Singapore (NUS). I am broadly interested in Trustworthy AI, LLM Reasoning and Agents. I am fortunate to work closely with Prof. Bryan Hooi and Prof. Tat-Seng Chua from NUS, Prof. Tianwei Zhang from NTU, Prof. Junxian He from HKUST, Prof. Muhao Chen from UC Davis, and Prof. Kai-Wei Chang from UCLA.

My primary research interests include:

Trustworthy AI: Hallucination Detection, Calibration & Adversarial Robustness.
AI Reasoning: Enhancing reasoning abilities of LLMs.
LLM Applications: Autonomous agents & RAG systems.

Email / Google Scholar / LinkedIn / Github / WeChat

Looking for Fall 2026 PhD opportunities on Trustworthy AI, LLM Reasoning and Agents, feel free to contact!

🔥 News

[2025.08] Two first-author papers accepted to EMNLP 2025! One to Main Track and one to Findings!
[2025.06] Our paper GuardReasoner-VL is accepted to ICML 2025 R2-FM Workshop!
[2025.04] Our survey on LRMs Safety is on arxiv now, check out the paper and repo!
[2025.01] One first-author paper is accepted to NAACL 2025 Main Conference.
[2025.01] I started my internship at Tiktok as an Algorithm Engineer Intern.
[2024.11] One first-author paper is accepted to COLING 2025.

📑 Pre-prints & Publications

^* denotes equal contribution.

	Unlocking the Pre-Trained Model as a Dual-Alignment Calibrator for Post-Trained LLMs Cheng Wang^, Beier Luo^, Hongxin Wei, Yixuan Li, Xuefeng Du Under Review, 2025
	Mirage or Method? How Model-Task Alignment Induces Divergent RL Conclusions Haoze Wu^, Cheng Wang^, Wenshuo Zhao, Junxian He Under Review, 2025 Paper / Code
	Neurips MechInterp Workshop (Spotlight) False Sense of Security: Why Probing-based Malicious Input Detection Fails to Generalize Cheng Wang^, Zeming Wei^, Qin Liu, Wenxuan Zhou, Muhao Chen Paper / Code
	Taming Extreme Tokens: Covariance-Aware GRPO with Gaussian-Kernel Advantage Reweighting Cheng Wang, Qin Liu, Wenxuan Zhou, Muhao Chen Under Review, 2025
	EMNLP 2025 MainWhen Audio and Text Disagree: Benchmarking Text Bias in Large Audio-Language Models under Cross-Modal Inconsistencies Cheng Wang, Gelei Deng, Xianglin Yang, Tianwei Zhang Paper / Code
	NeurIPS 2025GuardReasoner-VL: Safeguarding VLMs via Reinforced Reasoning Yue Liu, Shengfang Zhai, Mingzhe Du, Yulin Chen, Tri Cao, Hongcheng Gao, Cheng Wang, Xinfeng Li, Kun Wang, Junfeng Fang, Jiaheng Zhang, Bryan Hooi Paper / Code
	EMNLP 2025 FindingsSafety in Large Reasoning Models: A Survey Cheng Wang^*, Yue Liu, Baolong Bi, Duzhen Zhang, Zhongzhi Li, Junfeng Fang, Bryan Hooi Paper / GitHub
	NAACL 2025 MainTricking Retrievers with Influential Tokens: An Efficient Black-Box Corpus Poisoning Attack Cheng Wang, Yiwei Wang, Yujun Cai, Bryan Hooi Paper
	COLING 2025Con-ReCall: Detecting Pre-training Data in LLMs via Contrastive Decoding Cheng Wang, Yiwei Wang, Bryan Hooi, Yujun Cai, Nanyun Peng, Kai-Wei Chang Paper / Code

🎓 Education

National University of Singapore (NUS)
Period: 2022 - Present
Major: Computer Science & Math

💼 Professional & Industry Experience

	Tiktok \| Singapore Algorithm Engineer Intern Period: Jan 2025 - June 2025
	National University of Singapore \| Singapore Teaching Assistant, Introduction to AI and Machine Learning Period: Jan 2024 - May 2024

Last update: Sep 2025