About Me

 

I’m on the research team at Omnifold, where I solve forecasting and optimization problems with AI. Before Omnifold, I was a research scientist at Meta working on recommender systems.

I received my Ph.D. in Computer Science from UC Santa Barbara in 2022, advised by Professor William Y. Wang in the UCSB NLP Group. My Ph.D. research covered natural language processing and deep learning, in particular Bayesian deep learning, uncertainty estimation, deep generative models and their applications.

I received my M.S. in Data Science from NYU in 2016. I also hold a Bachelor’s from Tsinghua University and a Master’s degree from UC Davis both in Civil Engineering.

---

Publications & Preprints

 

On Hallucination and Predictive Uncertainty in Conditional Language Generation
Yijun Xiao, William Yang Wang
EACL 2021   [Paper]

Disentangled Representation Learning with Wasserstein Total Correlation
Yijun Xiao, William Yang Wang
arXiv:1912.12818   [Paper]

Quantifying Uncertainties in Natural Language Processing Tasks
Yijun Xiao, William Yang Wang
AAAI 2019   [Paper]

Dirichlet Variational Autoencoder for Text Modeling
Yijun Xiao, Tiancheng Zhao, William Yang Wang
arXiv:1811.00135   [Paper]

Elastic Responding Machine for Dialog Generation with Dynamically Mechanism Selecting
Ganbin Zhou, Ping Luo, Yijun Xiao, Fen Lin, Bo Chen, Qing He
AAAI 2018   [Paper]

Tree-Structured Neural Machine for Linguistics-Aware Sentence Generation
Ganbin Zhou, Ping Luo, Rongyu Cao, Yijun Xiao, Fen Lin, Bo Chen, Qing He
AAAI 2018   [Paper]

Efficient Character-level Document Classification by Combining Convolution and Recurrent Layers
Yijun Xiao, Kyunghyun Cho
arXiv:1602.00367   [Paper] [Code]

---