Huu-Tien Dang
Email: dtienuet@gmail.com
Google scholar | GitHub | Twitter
I am currently a Ph.D. student at the School of Information Science, Japan Advanced Institute of Science and Technology (JAIST), working with Prof. Naoya Inoue. I obtained my M.S. in Information Science from JAIST. Before JAIST, I finished my B.S. in Computer Science at Vietnam National University, advised by Prof. Xuan-Hieu Phan. I had a great time as an AI Research Intern at FPT Software AI Center, advised by the awesome Hoang Thanh-Tung.
Perspective on the development of science
Scientific research is an activity in which we, motivated by intellectual curiosity and a desire for exploration, seek to understand truths about various phenomena surrounding us, how they happen, what causes them to happen.
Research interests
- Robust machine learning
- Alignment: unlearning
Selected publications
- Beyond Forgetting: Machine Unlearning Elicits Controllable Side Behaviors and Capabilities, Huu-Tien Dang, The-Hai Nguyen, Dinh Mai Phuong, Minh-Phuong Nguyen, Hoang Thanh-Tung, Le-Minh Nguyen, and Naoya Inoue; Preprint, [Code]
- Improving Chain-of-Thought for Logical Reasoning via Attention-Aware Intervention, Nguyen Minh Phuong, Huu-Tien Dang, and Naoya Inoue; Findings of the Association for Computational Linguistics: EACL 2026, [Code]
- RegMean++: Enhancing Effectiveness and Generalization of Regression Mean for Model Merging, The-Hai Nguyen, Huu-Tien Dang, Takeshi Suzuki, and Le-Minh Nguyen; Preprint, [Code]
- Improving LLM Unlearning Robustness via Random Perturbations, Huu-Tien Dang, Hoang Thanh-Tung, Anh Bui, Minh-Phuong Nguyen, Le-Minh Nguyen, and Naoya Inoue; Preprint, [Code]
- On Effects of Steering Latent Representation for Large Language Model Unlearning, Huu-Tien Dang, Trung-Tin Pham, Hoang Thanh-Tung, Naoya Inoue; The 39th Annual AAAI Conference on Artificial Intelligence (AAAI 25, Oral), [Code], [Slides], [Video]
- Class Based Influence Functions for Error Detection, Nguyen Duc-Thang, Hoang Thanh-Tung, Quan Tran, Huu-Tien Dang, Hieu Nguyen, Anh Dau, Nghi Bui; The 61st Annual Meeting of the Association for Computational Linguistics (ACL 23), [Code], [Slides], [Video]
Program committee
Education
- School of Information Science, JAIST
Ph.D. student in Information Science - School of Information Science, JAIST
M.S. in Information Science
- Vietnam National University
B.S. in Computer Science
Experiences
- JAIST, Research Assistant, 10/2023 - present
- KiotViet, Software Engineer, 10/2022 - 9/2023
- FPT.AI (Engineer 04-10/2022); FPT Software AI Center (Research Intern; 10/2022 - 05/2023)
- Math and Science Summer Program (MaSSP), Mentor, 2022
- VinBigData, NLP Engineer, 01/2021 - 01/2022
- Institute of Mathematics, Vietnam Academic of Science and Technology, Research Intern, 2020
Honors and awards
- Science and Technology Scholarship from Vingroup Innovation Foundation