My vision for research is to develop an artificial intelligence (AI) agent that processes any type of input and produces any desired type of output. Such an agent can perceive surroundings with sensors, learn common sense with a knowledge base in English, listen to a spoken request in German, combine information from the different input sources and show the best answer in Korean text. To achieve this goal, I study the interaction between various languages/modalities and knowledge transfer across diverse tasks/domains, using the principles of machine learning (ML) and natural language processing (NLP).

Machine Translation

Speech Translation

Multimodal Learning

Machine Learning