Binh-Nguyen Nguyen

prof_pic.jpg

I am a second-year PhD Student in the Department of Data Science, New Jersey Institute of Technology, advised by Hai Phan. Previously, I was a research intern at A*STAR Agency for Science, Technology and Research, Singapore, where I was advised by Yang He.

I completed my Bachelor’s Degree in Computer Science at Vietnam National University, Hanoi (2021-2025). I first started doing research here, where I was advised by Hoang-Quynh Le and Duy-Cat Can.

My work focuses on improving the safety and efficiency of LLMs and agentic systems. Particularly, I am interested in and working on these directions:

  • Weak-to-strong distillation in LLMs with data privacy motivations.
  • Weight space learning for LLMs.
  • Reliable and personalized self-improving agents.

You can reach me at nbn4 [at] njit [dot] edu.

selected publications

  1. ICML
    Gradient Transformer: Learning to Generate Updates for LLMs
    Binh-Nguyen Nguyen, Khang Tran, NhatHai Phan, and Issa Khalil
    Forty-third International Conference on Machine Learning, 2026
    PDF
  2. COLING
    Swift Cross-Dataset Pruning: Enhancing Fine-Tuning Efficiency in Natural Language Understanding
    Binh-Nguyen Nguyen and Yang He
    The 31st International Conference on Computational Linguistics, 2025
    PDF
  3. JBI
    A multifaceted approach to drug-drug interaction extraction with fusion strategies
    Ba-Hoang Tran, Hung-Manh Hoang, Binh-Nguyen Nguyen, Duy-Cat Can, and Hoang-Quynh Le
    Journal of Biomedical Informatics, 2025
    PDF