Shuai Zhao

College of Computing and Data Science
Nanyang Technological University, Singapore, 639798.
Email: shuai.zhao αt ntu dοt edu dοt sg

📚 Google Scholar 📄 Full Publications Citations: 1428
Nanyang Technological University
Shuai Zhao
About

Biography

Hi! I am currently a Postdoctoral Researcher at Nanyang Technological University in Singapore, advised by Prof. Anh Tuan Luu. I received my PhD from Jinan University (2021–2024) under the supervision of Prof. Jinming Wen. I spent one year at Jinan University and was jointly trained at Nanyang Technological University for the remaining period, under the supervision of Prof. Anh Tuan Luu.

🤔My research interests include Model Security, Agent, Backdoor Attack, Large Language Models. I keep collobration with my friends Dr. Yanhao Jia on Multi-Modal Tasks.

🤝Note: I am always interested in working with self-motivated students. If you have a strong interest in large language model and model security, please feel free to contact me.

Updates

News

  • [2026.09]  🎉🎉 One paper accepted at ACM Computing Surveys.
  • [2026.09]  🎉🎉 One paper accepted at Pattern Recognition.
  • [2026.08]  🎉🎉 Two papers accepted at EMNLP2026, Congrats to Kaixuan Wang and Long Tuan Ho .
  • [2026.07]  🎉🎉 One paper accepted at IEEE Transactions on Information Forensics and Security.
  • [2026.06]  🎉🎉 One paper accepted at TPAMI, Congrats to Leilei Gan .
  • [2026.05]  🎉🎉 One paper accepted at ICML2026, Congrats to Xuan Feng .
  • [2026.04]  🎉🎉 Five papers accepted at ACL2026.
  • [2026.02]  🎉🎉 One paper accepted at IEEE Transactions on Dependable and Secure Computing.
  • [2026.01]  🎉🎉 One paper accepted at IEEE Transactions on Affective Computing.
  • [2025.12]  🎉🎉 One paper accepted at BMJ Open Ophthalmology.
  • [2025.12]  🎉🎉 One paper accepted at Pattern Recognition.
  • [2025.11]  🎉🎉 One paper accepted at IEEE Transactions on Information Forensics and Security.
  • [2025.11]  🎉🎉 Two papers accepted at AAAI2026.
  • [2025.11]  🎉🎉 One paper accepted at IEEE Transactions on Affective Computing.
  • [2025.10]  🎉🎉 Invited to serve as Area Chair for ARR; serve as Academic Editor for IET Biometrics Journal.
  • [2025.07]  🎉🎉 One paper accepted at Pattern Recognition.
  • [2025.05]  🎉🎉 One paper accepted at Information Fusion.
  • [2025.05]  🎉🎉 Three papers accepted at ACL2025.
  • [2025.04]  🎉🎉 One paper accepted at IEEE Transactions on Affective Computing.
  • [2025.04]  🎉🎉 One paper accepted at IJCNN2025.
  • [2025.02]  🎉🎉 One paper accepted at CVPR2025 (Highlight).
  • [2025.01]  🎉🎉 One paper accepted at NAACL2025.
  • [2025.01]  🎉🎉 One paper accepted at Transactions on Machine Learning Research (Survey Certification).
  • [2024.11]  🎉🎉 One paper accepted at Expert Systems With Applications.
  • [2024.09]  🎉🎉 One paper accepted at EMNLP2024.
  • [2024.05]  🎉🎉 One paper accepted at IEEE-ACM Transactions on Audio, Speech and Language Processing.
  • [2024.04]  🎉🎉 One paper accepted at IEEE Transactions on Emerging Topics in Computational Intelligence.
  • [2024.03]  🎉🎉 One paper accepted at NAACL2024.
  • [2023.10]  🎉🎉 One paper accepted at EMNLP2023.
  • [2023.05]  🎉🎉 One paper accepted at ACM Transactions on Asian and Low-Resource Language Information Processing.
Research Output

Selected Publications

First-author and corresponding-author publications
Full Publications
2026 2025 2024 2023 2022 and before

2026

  1. Jinming Wen, Xinyi Wu, Yuwen Li, Yanhao Jia, Shuai Zhao✉. Investigating Vulnerabilities and Defenses in Audio-Visual Attacks: A Survey on Multimodal Models. ACM Computing Surveys 2026. (Corresponding Author)
  2. Shuai Zhao, Zhongliang Guo, Luwei Xiao, Cong-Duy Nguyen, Yanhao Jia, Xinyi Wu, Luu Anh Tuan. Exploring Backdoor Attack and Defense Algorithms in LLMs: Enhancing Incontext Learning Security. Pattern Recognition 2026.
  3. Long Tuan Ho, Yahui Ding, Huimin Xu, Shuai Zhao✉, Anh Tuan Luu. You Can’t Trust Your Agent: A Survey on Backdoor Threats and Defenses in Intelligent Agents. EMNLP 2026. (Corresponding Author)
  4. Kaixuan Wang, Chenxin Diao, Jason T. Jacques, Zhongliang Guo, Loraine Clarke, Shuai Zhao✉. HarmReduction: Benchmarking LLMs in Harm Reduction Information Provision to Support People Who Use Drugs. EMNLP 2026. (Corresponding Author)
  5. Shuai Zhao, Zhongliang Guo, Xinyi Wu, Xiaobao Wu, Yanhao Jia, Luwei Xiao, Anh Tuan Luu. Protecting Your Customized LLM Systems from Backdoored Instructions with Metacognitive Probing. IEEE Transactions on Information Forensics and Security 2026.
  6. Shuai Zhao, Xinyi Wu, Shiqian Zhao, Xiaobao Wu, Zhongliang Guo, Yanhao Jia, Anh Tuan Luu. P2P: A Poison-to-Poison Remedy for Reliable Backdoor Defense in LLMs. ACL 2026.
  7. Man Hu, Xinyi Wu, Zhufeng Suo, Jinbo Feng, Linghui Meng, Yanhao Jia, Anh Tuan Luu, Shuai Zhao✉. Rethinking Reasoning: A Survey on Reasoning-based Backdoors in LLMs. ACL 2026. (Corresponding Author)
  8. Shuai Zhao, Qika Lin, Yanhao Jia, Xinyi Wu, Yuwen Li, and Luu Anh Tuan. UniFLE: Uniform Fusion of Multiple LoRA Experts for Backdoor Defense in Large Language Models. IEEE Transactions on Dependable and Secure Computing 2026.
  9. Xinyi Wu, Jia, Yanhao, Qinglin Zhang, Yiran Qin, Luwei Xiao, and Shuai Zhao✉. Towards Affective Evaluation of STEM Education: Leveraging MLLMs in Project-Based Learning. IEEE Transactions on Affective Computing 2026. (Corresponding Author)

2025

  1. Yulin Zhang†, Shuai Zhao†, Jianbing Ren, Yuwen Li, Xinyu Zhao, Chuan Nie, Suzhen Xie, Xuelin Huang, Jie Sun, Jinming Wen, Xianqiong Luo, Guoming Zhang. NLP-ROPCare: Predicting retinopathy of prematurity with admission notes using natural language processing[J]. BMJ Open Ophthalmology, 2025. (Equal contributions)
  2. Yuwen Li, Xinyi Wu, Zhongliang Guo✉, Luwei Xiao, Yanhao Jia, and Shuai Zhao✉. Backdoor Defense for Large Language Models with Weak-to-Strong Knowledge Distillation[J]. Pattern Recognition, 2025. (Corresponding Authors)
  3. Shuai Zhao, Yulin Zhang, Luwei Xiao, Xinyi Wu, Yanhao Jia, Zhongliang Guo, Xiaobao Wu, Cong-Duy Nguyen, Guoming Zhang, and Anh Tuan Luu. Affective-roptester: Capability and bias analysis of llms in predicting retinopathy of prematurity[J]. IEEE Transactions on Affective Computing, 2025.
  4. Guo, Zhongliang, Chun Tong Lei, Lei Fang, Shuai Zhao✉, Yifei Qian, Jingyu Lin, Zeyu Wang, Cunjian Chen, Ognjen Arandjelović, and Chun Pong Lau✉. A grey-box attack against latent diffusion model-based image editing by posterior collapse[J]. IEEE Transactions on Information Forensics and Security, 2025. (Corresponding Authors)
  5. Man, Hu, Yahui Ding, Yatao Yang✉, Liangyu Chen, Yanhao Jia, and Shuai Zhao✉. DUP: Detection-guided Unlearning for Backdoor Purification in Language Models[C]. AAAI, 2026. (Corresponding Authors)
  6. Yanhao Jia, Xinyi Wu, Hao Li, Qinglin Zhang, Yuxiao Hu, Shuai Zhao✉, Wenqi, Fan. Uni-Retrieval: A Multi-Style Retrieval Framework for STEM's Education[C]. ACL, 2025. (Corresponding Author)
  7. Shuai Zhao, Xiaobao Wu, Cong-Duy Nguyen, Yanhao Jia, Meihuizi Jia, Yichao Feng, Luu Anh Tuan. Unlearning Backdoor Attacks for LLMs with Weak-to-Strong Knowledge Distillation[C]. ACL, 2025.
  8. Shuai Zhao, Meihuizi Jia, Zhongliang Guo, Leilei Gan, Xiaoyu Xu, Xiaobao Wu, Jie Fu, Yichao Feng, Fengjun Pan, Luu Anh Tuan. A Survey of Recent Backdoor Attacks and Defenses in Large Language Models[J]. Transactions on Machine Learning Research, 2025.（Survey Certification）

2024

  1. Shuai Zhao, Xiaoyu Xu, Luwei Xiao, Jinming Wen, Luu Anh Tuan. Clean-label Backdoor Attack and Defense: An Examination of Language Model Vulnerability[J]. Expert Systems With Applications, 2024.
  2. Shuai Zhao, Meihuizi Jia, Anh Tuan Luu, Fengjun Pan, Jinming Wen. Universal Vulnerabilities in Large Language Models: Backdoor Attacks for In-context Learning[C]. Proceedings of the Empirical Methods in Natural Language Processing (EMNLP), 2024.
  3. Shuai Zhao, Anh Tuan Luu, Jie Fu, Jinming Wen, Weiqi Luo. Exploring Clean Label Backdoor Attacks and Defense in Language Models[J]. IEEE-ACM Transactions on Audio, Speech and Language Processing, 2024.
  4. Shuai Zhao, Jie Tian, Jie Fu, Jie Chen, Jinming Wen. FeaMix: Feature Mix with Memory Batch Based on Self-consistency Learning for Code Generation and Code Translation[J]. IEEE Transactions on Emerging Topics in Computational Intelligence, 2024.
  5. Shuai Zhao, Leilei Gan, Anh Tuan Luu, Jie Fu, Lingjuan Lv, Miehuizi Jia, Jinming Wen. Defending Against Weight-Poisoning Attacks for Parameter-Efficient Fine-Tuning, NAACL, 2024

2023

  1. Shuai Zhao, Jinming Wen, Anh Tuan Luu, Junbo Zhao, Jie Fu. Prompt as Triggers for Backdoor Attack: Examining the Vulnerability in Language Models[C]. Proceedings of the Empirical Methods in Natural Language Processing (EMNLP), 2023.
  2. Shuai Zhao, Qing Li, Yuer Yang, Jinming Wen, Weiqi Luo. From Softmax to Nucleusmax: A Novel Sparse Language model for Chinese Radiology Report Summarization[J]. ACM Transactions on Asian and Low-Resource Language Information Processing, 2023.

2022 and before

  1. Shuai Zhao, Zhuoqian Liang, Jinming Wen, Jie Chen. Sparsing and smoothing for the seq2seq models[J]. IEEE Transactions on Artificial Intelligence, 2022: 1-10.
  2. Shuai Zhao, Qing Li, Tengjiao He, Jinming Wen. A Step-by-Step Gradient Penalty with Similarity Calculation for Text Summary Generation[J]. Neural Processing Letters, 2022: 1-16.
  3. Shuai Zhao, Tengjiao He, Jinming Wen. Sparse summary generation[J]. Applied Intelligence, 2023, 53(2): 2042-2054.
  4. Shuai Zhao, Tianyu Zhang, Man Hu, Wen Chang, Fucheng You. AP-BERT: enhanced pre-trained model through average pooling[J]. Applied Intelligence, 2022, 52: 15929–15937.
  5. Shuai Zhao, Fucheng You, Wen Chang, Tianyu Zhang, Man Hu. Augment BERT with average pooling layer for Chinese summary generation[J]. Journal of Intelligent & Fuzzy Systems, 2022, 42(3): 1859-1868.
  6. Zihao He, Yuer Yang, Shuai Zhao✉. Towards Pre-trained Language Model for Dynamic Disturbance[J]. The 3rd International Academic Exchange Conference on Science and Technology Innovation (IAECST), 2021, 480-484.
Systems

Projects

📌 Project 1 — Policy QA System (RAG-based)

Developed a domain-specific question answering system for public policy consultation, enabling users to retrieve accurate and up-to-date policy information through natural language queries.

  1. Designed document ingestion pipeline for structured and unstructured policy data
  2. Implemented retrieval optimization (chunking, embedding tuning, reranking)
  3. Improved answer factuality and reduced hallucination via grounded generation

🎓 Project 2 — University AI Assistant (NTU-GPT)

Designed and implemented a university-level AI assistant to support students and staff with academic, administrative, and campus-related queries.

  1. LLM-based responses for general queries
  2. RAG-based retrieval for factual / institutional knowledge

🧑‍🏫 Project 3 — Persona-based Conversational Agent

Created a persona-driven chatbot that simulates the speaking style and perspectives of a historical figure for educational and interactive purposes.

  1. Designed prompt engineering strategy to maintain persona consistency
  2. Curated and structured historical texts for grounding responses
  3. Combined LLM generation with retrieval to ensure factual alignment
Funding

Academic Projects

Background

Education and Professional Experience

Recognition

Selected Honors and Awards

Community

Academic Service