Zuchao Li

Email: charlee [AT] sjtu.edu.cn


Zuchao Li (Charlie) is a Ph.D. student in the Department of Computer Science and Engineering at Shanghai Jiao Tong University, advised by Prof. Hai Zhao.

He was an internship research fellow at NICT from 2019-2020, working with Dr. Rui Wang, Masao Utiyama, and Eiichiro Sumita. He received his B.S. degree from Wuhan University in 2017. His research focuses on Syntactic Parsing, Semantic Parsing and Language Modeling, especially the Neural Machine Translation. He (as the first author) has published 9 papers in leading NLP/ML/AI conferences and journals, including ACL, EMNLP, COLING, ICLR, AAAI, PRICAI and IEEE/ACM transactions. He (as the first accomplisher) has won the 1st place in various Parsing and NMT shared tasks, such as CoNLL-2019, WMT-2020. He served as a program committee member of ACL, EMNLP, COLING, ICLR, AAAI, and CCL.

News & Memorabilia


Syntactic Parsing

Memory Network for Linguistic Structure Parsing
Global Greedy Dependency Parsing
Cross-Domain Transfer Learning for Dependency Parsing
Effective Representation for Easy-First Dependency Parsing
Joint Learning of POS and Dependencies for Multilingual Universal Dependency Parsing
Seq2seq Dependency Parsing

Semantic Parsing (*: as the co-first author)

Neural Unsupervised Semantic Role Labeling
High-order Semantic Role Labeling
Parsing All: Syntax and Semantics, Dependencies and Spans
Adaptive Convolution for Semantic Role Labeling
SJTU-NICT at MRP 2019: Multi-Task Learning for End-to-End Uniform Semantic Graph Parsing
Syntax-aware Multilingual Semantic Role Labeling
Dependency or Span, End-to-End Uniform Semantic Role Labeling
A Unified Syntax-aware Framework for Semantic Role Labeling
A Full End-to-End Semantic Role Labeler, Syntactic-agnostic or Syntactic-aware?
Syntax for Semantic Role Labeling, to Be, or Not to Be

Language Modeling

Text Compression-aided Transformer Encoding
SJTU-NICT's Supervised and Unsupervised Neural Machine Translation Systems for the WMT20 News Translation Task
Reference Language based Unsupervised Neural Machine Translation
Data-dependent Gaussian Prior Objective for Language Generation
Neural Machine Translation with Universal Visual Representation
Explicit Sentence Compression for Neural Machine Translation
Semantics-aware BERT for Natural Language Understanding
Explicit Contextual Semantics for Text Comprehension
Effective Subword Segmentation for Text Comprehension

Demo (*: as the co-first author)

Neural-based Chinese Pinyin Aided Input Method with Customizable Association


High precision gesture sensing via quantitative characterization of the Doppler effect