NTU AIS Lab

National Taiwan University Artificial Intelligence Security Lab

臺灣大學 人工智慧安全實驗室 陳尚澤教授

⚖️Welcome to NTU AIS Lab

Welcome to the National Taiwan University Artificial Intelligence Security Lab. Our research focuses on trustworthy machine learning, emphasizing 🛡️robustness, 🔒privacy, and ⚖️fairness. While the accuracy of machine learning models in many fields is already very high, often surpassing human performance, they are increasingly being used in critical human-related decision-making processes. However, accuracy alone is not enough for practical applications. The robustness, reliability, security, and fairness of machine learning models are equally crucial. In our research, we aim to identify vulnerabilities in machine learning models under various environments and assumptions and propose solutions to strengthen these models. Our goal is to create reliable, secure, and fair machine learning systems, thereby fostering greater public trust in the technology.

We are always looking for highly motivated undergraduate and graduate students. Please contact us if you are interested in our research.

This is our twitch channel: AISLab. Follow us for random stuff

⚖️News

  • Sep. 2026 - Two papers accepted to NeurIPS 2026.
  • Jun. 2026 - One paper accepted to ECCV 2026.
  • Jan. 2026 - One paper accepted to EACL 2026.
  • May 2025 - Two papers accepted to ICML 2025.
  • Jan. 2025 - One paper accepted to NACCL-Findings 2025.
  • Nov. 2024 - MS student Jacky Liu won 中華民國人工智慧學會 TAAI 2024 碩士論文佳作獎!
  • Sep. 2024 - One paper accepted to NeurIPS 2024.
  • Sep. 2024 - One paper accepted to EMNLP 2024.
  • Jun. 2024 - Congratulations to our undergrad members for getting the awards in 大學部專題展. Pin-Han Huang got the first prize & the Appier prize, Ren-Wei Liang and Weiping Li got the the Vysioneer award.
  • Mar. 2024 - One paper accepted to CVPR Workshop 2024.
  • Jan. 2024 - One paper accepted to ICLR 2024.
  • Dec. 2023 - One paper accepted to AAAI 2024.
  • Jun. 2023 - Congratulations to our undergrad members Jan-Ting and Yu-Syun for getting the awards in 大學部專題展.
  • Nov. 2022 - Jacky Liu receives Garmin Fellowship. Congrats!
  • Aug. 2021 - Prof. Chen co-organized MLSS 2021.
  • Jul. 2021 - Hung-Jui and Wei-Lun got the Appier AI Potential Award and the Honorable Mention Award at the NTU CSIE Undergraduate Research Competition, respectively. Congrats!

⚖️Our Team

Advisor

👑Prof. Shang-Tse Chen

PhD Students

👨‍🎓Bo-Han Kung       👨‍🎓Bo-Han Lai       👨‍🎓Teng-Yu Lin       👨‍🎓Yin-Chieh Wang

MS Students

👨‍🎓Sheng-Yuan Chang      👨‍🎓Kin-Fong Chao      👨‍🎓Jia-You Chen      👨‍🎓Ting-Chi Chen      👨‍🎓Kai-Yu Cheng     
👨‍🎓Chien-Jui Huang      👨‍🎓Jie-Ling Lee      👨‍🎓Ren-Wei Liang      👨‍🎓Jay Chiehen Liao      👨‍🎓Bo-Wei Lin     
👨‍🎓Yu-Sheng Ma      👨‍🎓Sih-Ray Mao      👨‍🎓Ming-Jun Yang

Cooperator

👨‍🎓Long-Hin Fung

⚖️Alumni

PhD Alumni

🎓Hsuan Su

MS Alumni

🎓Yu-Cheng Cheng      🎓Hung-Yeh Chien      🎓Yu-Ling Hsu      🎓Yu-Che Huang      🎓Edward Lee     
🎓Kuan-Hsun Li      🎓Chung-Hao Liao      🎓Zhen-Ting Liu      🎓Jun-Jie Wang      🎓Hung-Yang Yeh

⚖️Publications

Robust Concept Unlearning

Robust Concept Unlearning in Diffusion Models via Directional Stability Regularization

Bo-Han Lai, Hsuan-Tien Lin, Chia-Mu Yu, Han Zhao, and Shang-Tse Chen

NeurIPS'26
DiffRep

Expanding the Role of Diffusion Models for Robust Classifier Training

Pin-Han Huang, Shang-Tse Chen, and Hsuan-Tien Lin

NeurIPS'26
ReTarget

ReTarget: Representation Transformation via Adversarial Regularization for Geometric Misalignment

Jia-You Chen and Shang-Tse Chen

ECCV'26
Preference Vector

Adaptive Helpfulness–Harmlessness Alignment with Preference Vectors

Ren-Wei Liang, Chin-Ting Hsu, Chan-Hung Yu, Saransh Agrawal, Shih-Cheng Huang, Chieh-Yen Lin, Shang-Tse Chen, Kuan-Hao Huang, and Shao-Hua Sun

EACL'26
bronet

Enhancing Certified Robustness via Block Reflector Orthogonal Layers and Logit Annealing Loss

Bo-Han Lai*, Pin-Han Huang*, Bo-Han Kung*, Shang-Tse Chen

ICML'25 (spotlight)
DRAG

DRAG: Data Reconstruction Attack using Guided Diffusion

Wa-Kin Lei, Jun-Cheng Chen, and Shang-Tse Chen

ICML'25
JUMP

Jailbreaking with Universal Multi-Prompts

Yu-Ling Hsu, Hsuan Su, and Shang-Tse Chen

NAACL-Findings'25
Trap-MID

Trap-MID: Trapdoor-based Defense against Model Inversion Attacks

Zhen-Ting Liu and Shang-Tse Chen

NeurIPS'24
Annealing

Annealing Self-Distillation Rectification Improves Adversarial Training

Yu-Yu Wu, Hung-Jui Wang, and Shang-Tse Chen

ICLR'24
QCRS

Towards Large Certified Radius in Randomized Smoothing using Quasiconcave Optimization

Bo-Han Kung and Shang-Tse Chen

AAAI'24
AdvCAPTCHA

Enhancing Targeted Attack Transferability via Diversified Weight Pruning

Hung-Jui Wang, Yu-Yu Wu, and Shang-Tse Chen

CVPR workshop'24
AdvCAPTCHA

AdvCAPTCHA: Creating Usable and Secure Audio CAPTCHA with Adversarial Machine Learning

Hao-Ping (Hank) Lee, Wei-Lun Kao, Hung-Jui Wang, Ruei-Che Chang, Yi-Hao Peng, Fu-Yin Cherng, and Shang-Tse Chen

USEC'24
Ensuring

Ensuring Bidirectional Privacy on Wireless Split Inference Systems

Chia-Che Sa, Li-Chen Cheng, Hsing-Huan Chung, Te-Chuan Chiu, Chih-Yu Wang, Ai-Chun Pang, and Shang-Tse Chen

IEEE Wireless Communications Magazine 2024
Position

Position Matters! Empirical Study of Order Effect in Knowledge-grounded Dialogue

Hsuan Su, Shachi H Kumar, Sahisnu Mazumder, Wenda Chen, Ramesh Manuvinakurike, Eda Okur, Saurav Sahay, Lama Nachman, Shang-Tse Chen, and Hung-yi Lee

ACL DialDoc Workshop'23
Fair

Fair Robust Active Learning by Joint Inconsistency

Tsung-Han Wu, Hung-Ting Su, Shang-Tse Chen, and Winston H. Hsu

ICCV Workshop'23

⚖️FAQ

  • Q: How can I join the NTU AIS Lab?
    A: We welcome self-motivated students, including undergraduates and graduates, to join us. If you are interested in the research topics of the lab, please email Professor Chen and include a brief resume. Professor Chen will arrange a short interview.
  • Q: What are the research areas of the NTUAIS Lab?
    A: Our research focuses on trustworthy machine learning, including robustness, privacy, and fairness.
  • Q: How can I learn more about the NTU AIS Lab's research?
    A: You can explore our publications and projects on this website.
  • Q: I am an undergraduate student. When is the best time for me to join the lab for a project?
    A: The sooner, the better.
  • Q: What qualities or skills does the professor expect from students working on their projects?
    A: The professor expects students to have a genuine desire to learn, be proactive rather than passive, and actively explore, with sufficient background in linear algebra and probability, and ideally, having completed a course in machine learning.

⚖️Contact Us

Prof. Shang-Tse Chen

Email: stchen@csie.ntu.edu.tw

Website: https://www.csie.ntu.edu.tw/~stchen/

Professor's Office: NTU CSIE Building 515

Lab Location: NTU CSIE Building 405

Join us !