Master's student at Mila – Quebec AI Institute and University of Montreal.
-🌱 Previous journey: AI Research Resident at FPT Software AI Center (AIC), AI Engineer at Data & AI Lab (DAL), VNG Corporation.
-
Multimodal Models Reasoning & Understanding: Multimodal Alignment, Representation Learning, Structured Representation, Compositional Reasoning, and Fine-grained Understanding.
-
Resource-efficient Models: Parameter-Efficient Fine-Tuning (PEFT) (Efficient Training), Distilled/Small Models (Efficient Inference), Token Merging/Pruning (Efficient Input).
-
Multimodal Models Generation: Embodied Agent, Multimodal Chatbot, Visual Programming.
My current research experience comprises of Intelligent Surveillance Systems, Image/Video Understanding, Multimodal Learning and PEFT including:
-
[2023-Present] Efficient Cross-Modal Learning & Understanding: Video-Language Matching, Parameter-Efficient Fine-Tuning (PEFT), Multimodal Compositionality, Structured Representation (Scene Graph Generation).
-
[2021-2023] Intelligent Industrial/Traffic Systems Applications: Tracked-Vehicle to Video Retrieval, Person/Vehicle Re-Identification, Person/Vehicle Tracking, Face Recognition/Verification.