Welcome! I am an AI Researcher at Zalo AI & UIT@NLP at Ho Chi Minh city, Vietnam. I have been a researcher at UIT@NLP lab during my undergraduate degree at University of Information Technology. After receiving my bachelor’s degree in computer science, I have continued to do part-time research at this lab until now. In parallel, I have worked as an AI researcher at Zalo AI, Vietnam for two years.

Currently, I am also looking for an opportunity for a PhD position in the field of AI and NLP, especially Large Language Model.

Research Interests

I believe retrievers still play a significant role in the current era of AI development, especially the strong growth of generative AI. Retrievers are the way for models to access the most realistic, accurate, and recent data sources. My two favorite research directions right now are two problems revolving around information retrieval.

• How to reason effectively during retrieval? This question drives my research interests, particularly in exploring how to organize diverse data forms—such as text, images, audio, and code—to enable more effective reasoning processes for identifying the most relevant evidence for a query. For me, the Knowledge graph is a promising approach to handling the above task because of its structure. My goal is to integrate advanced reasoning mechanisms into this structure to represent and reason on it to find the evidence effectively and accurately.

• How to represent multilingual multimodal data through embedding? How can multilingual and multimodal data be effectively represented through embeddings? My goal is to develop methods that enable such representation, as current embedding models are often limited to specific data domains or languages without effectively bridging different aspects. I also see great potential in applying these methods to knowledge graphs, which could enhance the integration and retrieval of complex information.

I am happy to chat and discuss potential collaborations. Please feel free to reach out to me via Email (phongdntvn@gmail.com).

News

June 19, 2024:

• Phong Do published a new research paper titled “VLUE: A New Benchmark and Multi-task Knowledge Transfer Learning for Vietnamese Natural Language Understanding” at NAACL 2024 (Findings).