The vision of my research is to understand the underlying mechanism of human intelligence from both the neural and cogintive perspectives, and finally approach the grand goal of building Artificial General Intelligence (AGI). More specifically, I research this broad question: “how can we teach the machine to acquire language ability?”, also as known as Natural Language Processing (NLP).

I recently work on the crosslingual language generation which challenges recent pretrained language models on three fundamental aspects:

1. Systematic Generalization
2. Few-shot Learning
3. Learning for Relational Knowledge (e.g., physical-world knowledge)

I pursue my doctoral studies at the Institute of Language, Cognition and Computation at University of Edinburgh and Language Technology Lab at University of Cambridge, and I am fortunated to be advised by Shay Cohen, Anna Korhonen and Edoardo Ponti under ELLIS PhD & PostDoc program, a pan-European network of excellence in AI research. I am also a member of EdinburghNLP.

Prior to starting my PhD, I spent time at Baidu NLP as a Research Intern where I worked on Open-domain Question Answering and Dense Retrieval. I obtained my MSc from University of Edinburgh and worked on text summarization and structured prediction.

Download my resumé.