Email: thisisjcykcd AT gmail.com
- Aug. 2018 - Present
Ph.D. student, The Chinese University of Hong Kong
- Sept. 2015 - Mar. 2018
M.S., Dept. of Computer Science, Shanghai Jiao Tong University
- Sept. 2011 - Jun. 2015
B.E., Dept. of Computer Science, Xiamen University
- Jul. 2017 - Jan. 2018, Visiting scholar
Singapore University of Technology and Design, Advisor: Prof. Yue Zhang
- Feb. 2017 -Jun. 2017, Research intern
Tencent AI Lab, Mentors: Dr. Xiaojiang Liu, Dr. Shuming Shi
- Sept. 2015 -Feb. 2017, Research assistant
Shanghai Jiao Tong University, Supervisor: Prof. Hai Zhao
Selected Publications: (Google Scholar)
- Skeleton-to-Response: Dialogue Generation Guided by Retrieval Memory. [arxiv]
Deng Cai, Yan Wang, Wei Bi, Zhaopeng Tu, Xiaojiang Liu, Wai Lam, and Shuming Shi. To appear in Proceedings of the NAACL 2019 Conference, 2019. (NAACL 2019)
- Unsupervised Learning helps Supervised Neural Word Segmentation. [paper]
Xiaobin Wang, Deng Cai, Guangwei Xu, Hai Zhao, Linlin Li and Luo Si. In Proceedings of the AAAI 2019 Conference, 2019. (AAAI 2019)
- Translating a Math Word Problem to a Expression Tree. [paper]
Lei Wang, Yan Wang, Deng Cai, Dongxiang Zhang and Xiaojiang Liu. In Proceedings of the EMNLP 2018 Conference, 2018. (EMNLP 2018).
- Fast and Accurate Neural Word Segmentation for Chinese. [paper] [code]
Deng Cai, Hai Zhao, Zhisong Zhang, Yuan Xin, Yongjian Wu, and Feiyue Huang. In Proceedings of the ACL 2017 Conference, 2017. (ACL 2017).
- A Hybrid Model for Chinese Spelling Check. [paper]
Deng Cai*, Hai Zhao* (first authors), Yang Xin, Yuzhu Wang, Zhongye Jia. ACM Transactions on Asian Low-Resource Language Information Process, 2017. (TALLIP).
- Neural Word Segmentation Learning for Chinese. [paper] [code] [slides]
Deng Cai and Hai Zhao. In Proceedings of the ACL 2016 Conference, 2016. (ACL 2016).
Open Source Projects: (Github Profile)
- Fast and Accurate Neural Chinese Word Segmentation
Efficient and effective neural Chinese word segmentation using direct structured prediction approach. Works have been presented at ACL2016 and ACL2017.
I used to contribute actively to Dynet. A list of my contributions can be found here.
A simple yet complete implementation of the popular BERT model, tested on large-scale distributed environment (more than 80GPUs).
- Biaffine Dependency Parser
A Dynet implementation of the popular biaffine dependency parser, achieving state-of-the-art performance. Much simpler than the original tensorflow code. I played around many domain adaptation methods on this model.
- Implicit Discourse Relation Classification
Pair-aware sentence modeling for Implicit Discourse Relation Classification. A summary of this work: [paper]
Selected Awards and Honors:
- National Scholarship for Graduate Student (top 2% students), Ministry of Education of P.R.China, 2016
- Excellent Undergraduate Thesis Award, XMU, 2015
- Dean's List, 2014, School of Information Science and Engineering, XMU, 2014
- Gold Medal, The 5-th Fujian Provincial University Programming Contest, 2014
- Bronze Medal, ACM-ICPC Asia Regional Programming Contest, 2013
Reviewer (or PC Member): ACL(2017-2019), COLING(2016), EACL(2017), IJCNLP(2017), LREC(2018)
- What?! I remember clearly, last time I played the game, I was an international master.
- I cannot swim well because my density is too large.