People

Jinfeng Rao

Research Scientist

I joined Facebook Conversational AI team as a research scientist in September 2018, where I work on natural language generation and dialog. Before that, I was a visiting researcher at Stanford University. I obtained my CS PhD from University of Maryland College Park in 2018, and a bachelor’s degree from Zhejiang University in 2013.

During my PhD, I had some research internship experiences at Comcast AI Research, Google and Microsoft. At Comcast, I helped build their XFINITY voice search system with neural networks, which helped win the 69th Emmy Award (2017) for the technical contribution in advancing television technologies.

Interests

Deep learning, natural language processing and information retrieval

Latest Publications

Bridging the Gap Between Relevance Matching and Semantic Matching for Short Text Similarity Modeling

Jinfeng Rao, Linqing Liu, Yi Tay, Wei Yang, Peng Shi, Jimmy Lin

EMNLP - September 10, 2019

Lightweight and Efficient Neural Natural Language Processing with Quaternion Networks

Yi Tay, Aston Zhang, Luu Anh Tuan, Jinfeng Rao, Shuai Zhang, Shuohang Wang, Jie Fu, Siu Cheung Hui

ACL - August 1, 2019

Simple and Effective Curriculum Pointer-Generator Networks for Reading Comprehension over Long Narratives

Yi Tay, Shuohang Wang, Luu Anh Tuan, Jie Fu, Minh C. Phan, Xingdi Yuan, Jinfeng Rao, Siu Cheung Hui, Aston Zhang

ACL - August 1, 2019

Constrained Decoding for Neural NLG from Compositional Representations in Task-Oriented Dialogue

Anusha Balakrishnan, Jinfeng Rao, Kartikeya Upasani, Michael White, Rajen Subba

ACL - August 1, 2019

Multi-Perspective Relevance Matching with Hierarchical ConvNets for Social Media Search

Jinfeng Rao, Wei Yang, Yuhao Zhang, Ferhan Ture, Jimmy Lin

AAAI - May 3, 2019