I’m Minh-Duc, a young researcher with a strong background in mathematics and informatics. I earned my Bachelor’s degree from Hanoi University of Science and Technology (HUST), where I was guided by Dr. Tran Ngoc Thang in research on recommendation systems incorporating Multi-objective learning to enhance user-item matching.

After graduation, I joined the R2Studio reseach group on AI-empowered Search and Recommendation at VinUniversity, under the leadership of Assistant Professor Le Duy Dung (Andrew). Here, our team explores optimization algorithms for multi-objective optimization, contributing to advancements in this field.

Research Topics

• Multi-objective optimization (MOO), Multi Tasks Learning, Pareto Set Learning
• Large Language Models (LLMs) for Recommendation Systems (RS)
• AI in Applications

News

• [Feb.2026] I was accepted into the Master of Science in Computer Science program at VinUniversity for the 2025–2026 academic year.
• [Dec.2025] Paper “PromptHG: Prompt-Enhanced Heterogeneous Graph for Personalized News Recommendation” is accepted at the ECIR 2026 Full Papers track.
• [Jan. 2025] Our paper “Keyword-driven Retrieval-Augmented Large Language Models for Cold-start User Recommendations” is accepted at The Web Conference 2025, Workshop LLL 4 E-Commerce.
• [Dec. 2024] My first paper “Improving Pareto Set Learning for Expensive Multi-objective Optimization via Stein Variational Hypernetworks” is accepted at AAAI Conference on Artificial Intelligence (AAAI 2025).

Conference Papers

PromptHG: Prompt-Enhanced Heterogeneous Graph for Personalized News Recommendation
Hai-Dang Kieu, Delvin Ce Zhang, Minh Duc Nguyen, Qiang Wu, Min Xu, Dung D. Le
The 48th European Conference on Information Retrieval (ECIR 2026)

Enhancing News Recommendation with Hierarchical LLM Prompting
Hai-Dang Kieu, Delvin Ce Zhang, Minh Duc Nguyen, Min Xu, Qiang Wu, Dung D. Le
WWW ‘25: Companion Proceedings of the ACM on Web Conference 2025. Paper

Keyword-driven Retrieval-Augmented Large Language Models for Cold-start User Recommendations
Hai-Dang Kieu, Minh Duc Nguyen, Thanh-Son Nguyen, Dung D. Le
WWW ‘25: Companion Proceedings of the ACM on Web Conference 2025. Paper

Improving Pareto Set Learning for Expensive Multi-objective Optimization via Stein Variational Hypernetworks
Minh-Duc Nguyen, Phuong Mai Dinh, Quang-Huy Nguyen, Hoang Phi Long, Dung D. Le
Proceedings of the AAAI Conference on Artificial Intelligence, 39(18), 19677-19685. (AAAI25) Paper, Code

Interests

Football, Guitar, Books.

Gallery