About Me

Xin chΓ o vietnam icon ! I'm Huy, Research Master's student at Mila and University of Montreal, advised by Aishwarya Agrawal. I'm on my way to explore the world in every aspect through my own lens by becoming a researcher.

In my previous chapter, I earned my B.Eng. from International University – Vietnam National University HCMC in 2024. During my undergrad, I was fortunate to work as an AI Engineer at VNG Corporation and AI Research Resident at FPT Software AI Center. My undergrad team was one of the winners at CVPR 2023 NVIDIA AI City Challenge.

Recent Updates

  • Nov 2025: One paper is accepted at AAAI 2026πŸŽ‰πŸŽ‰
  • Oct 2025: I have been selected as NeurIPS 2025 Top Reviewer and awarded a complimentary registration to attend the conference!πŸŽ‰πŸŽ‰
  • Aug 2025: Bonjour MontrΓ©al πŸ‡¨πŸ‡¦! I will start my new journey at Mila and University of Montreal working on challenging and exciting problems in Multimodal Vision-Language domain!
  • Jul 2025: One paper is accepted at ACM MM 2025πŸŽ‰πŸŽ‰
  • Nov 2024: I was awarded a Student Registration Scholarship to attend ACCV 2024, hosted in my hometown, Hanoi! πŸ‡»πŸ‡³ ⛩️
  • Oct 2024: Finally, I graduated with a B.Eng from HCMIU! A huge thanks and deepest appreciation to my family, friends, and colleagues who supported me over the past four years and celebrated this special milestone with me! πŸŽ‰πŸŽ‰
  • Apr 2024: Our paper, titled WAVER: Writing-Style Agnostic Text-Video Retrieval via Distilling Vision-Language Models Through Open-Vocabulary Knowledge has been selected for an Oral Presentation at ICASSP 2024, Seoul, Korea! πŸ‡°πŸ‡· πŸŽ‰πŸŽ‰
  • Mar 2024: I will serve as a Session Chair for ICASSP 2024
  • Jan 2024: One paper is accepted at ICASSP 2024πŸŽ‰πŸŽ‰
  • Jun 2023: Our team at HCMIU simultaneously won the Winner Award πŸ† in Track 2 and the Runner-up Award πŸ₯ˆ in Track 1 at The 7th AI City Challenge Workshop, CVPR 2023, hosted by NVIDIA πŸŽ‰πŸŽ‰

Research Interests

My current research interests and works lie in generalizing and optimizing Foundation Multimodal Models πŸ‘€βœοΈπŸ€–πŸŒ:
Decision-making Agents: Reinforcement Learning, World Models, Unsupervised Exploration, Open-ended Learning.
Understanding & Reasoning: Alignment, Compositionality, Fine-grained and Structured Representation Learning.
Resource-efficient Methods: Token Compression, Parameter-efficient Modeling for Training & Inference.

Professional Services

Conference Reviewer: WACV (2026), NeurIPS (2025 - Top Reviewer), CoRL (2025), ICCV (2025), ACMMM (2025), ICASSP (2025)
Session Chair: ICASSP (2024)