KHOA TRÍ TUỆ NHÂN TẠO

Học viện Công nghệ Bưu chính Viễn thông

Mai Duc Binh

Teaching Assistant

Mai Duc Binh

Mai Duc Binh graduated with Excellent Honors from the Posts and Telecommunications Institute of Technology (PTIT), Hanoi, in 2026. He is currently a teaching assistant at the Faculty of Artificial Intelligence, PTIT, where he previously worked as a research assistant. His current research focuses on generative models, with particular interests in accelerating inference, reducing computational requirements, and enabling efficient deployment on resource-constrained hardware. He is also interested in applying generative models to real-time applications, where latency, computational efficiency, and resource utilization are critical considerations.

Generative models, speech processing
binhmd@ptit.edu.vn

Previous publications

No Article Authors Journal/Conference Year
1

Miximotion: One-step text-to-motion generation via asymmetric set distillation

H. Dinh, Binh Mai*, Q. B. L. Tran, L. Nguyen, and C. Tran

Under review at the International Symposium on Information and Communication Technology (SOICT), co-first author

 2026
2

Swiftaudio: Data-efficient caption-only distillation for one-step text-to-audio diffusion-based generation

Binh Mai, T. Q. B. Le, H. Dinh, and C. Tran

Under review at IEEE Transactions on Knowledge and Data Engineering.

 2026
3

Synthesizing cultural heritage: An end-to-end system for designing jewelry with vietnamese hue imperial motifs

H. Nguyen, Binh Mai*, H. Hoang, H. Dao, and C. Tran

Proceedings of the International Symposium on Information and Communication Technology (SOICT), Co-first author, Nha Trang, Vietnam

 2025
4

Ts2f: Text-assisted speech-to-face generation

Binh Duc Mai, H. M. Dinh, and C. Tran

ICT Express

 2025