Jiefu Ou
Hi there!
I am a first-year PhD student at the Center for Language and Speech Processing, Johns Hopkins University, advised by Daniel Khashabi and Benjamin Van Durme. Previously, I was a research master’s student (MLT) at Language Technologies Institute, Carnegie Mellon University, working with Teruko Mitamura and Daniel Fried.
Prior to CMU, I received my B.S. degree in Computer Science, and Mathematics & Economics from the Hong Kong University of Science and Technology. During my undergrad study, I spent two years at Knowledge Computation Group@HKUST, where I was advised by Yangqiu Song. I was also a research intern at CLSP, JHU supervised by Benjamin Van Durme.
Research Themes
I am broadly interested in Natural Language Processing. My (current) goal of research is to build intelligent systems that can communicate and collaborate with humans in dynamic environments for solving complex real-world tasks. Under this theme, I would like to explore:
- Language Grounding: How can we tie language to the surrounding world (e.g. multimodal context/embodied environment)? And by doing so, can we obtain the language representation that is aware of perception, knowledge, and dynamics of the world?
- Interaction: With the tremendous progress in the development and deployment of AI models (e.g. ChatGPT), how should we adapt these models to interact more effectively with human users?
I regard these two directions to be highly inter-connected and mutually beneficial: to use language as the mean of system-human interaction, grounding the communication to contexts helps the system understand user preference and intent so that to adjust its behavior accordingly to enable better collaboration.
In the past, I also worked on information extraction, controlled generation, discourse analysis, and knowledge graph completion.