Liang Pang

庞 亮

Associate Professor

Data Intelligence System Research Center

Institute of Computing Technology (ICT)

Chinese Academy of Sciences (CAS)

  • Email : pangliang AT
  • Office: : (8610)62600964
  • Address : NO. 6 Kexueyuan South Road, Haidian District, Beijing, P.R.China, 100190
  • Biography
  • Publication
  • ToolKits & Demos

About Me

Liang Pang is an associate professor at Institute of Computing Technology, Chinese Academy of Sciences. He received Ph.D. degree in Computer Science from University of Chinese Academy of Sciences in 2018 under supervising by Prof. Guojie Li. Before that, He get a B.S. degree in Software Engineering from Huazhong University of Science and Technology in 2012.

His current research is focused on deep learning in text matching tasks, including ranking, query answering, paraphrase identification.


SIGIR 2021 Tutorial

Liang Pang, Qingyao Ai, Jun Xu. Beyond Probability Ranking Principle: Modeling the Dependencies among Documents. The 44th International ACM SIGIR Conference on Research and Development in Information Retrieval. (SIGIR 2021) [pdf] [slides] [videos]

WSDM 2021 Tutorial

Liang Pang, Qingyao Ai, Jun Xu. Beyond Probability Ranking Principle: Modeling the Dependencies among Documents. The 14th ACM International Conference on Web Search and Data Mining. (WSDM 2021) [pdf] [slides] [Part1] [Part2] [Part3]

CIPS Summer School 2018 中国中文信息学会《前沿技术讲习班》

Jun Xu, Liang Pang. Deep and Reinforcement Learning for Information Retrieval.信息检索中的深度强化学习新进展. [slides] [site]

Open Source

AISO - Adaptive Information Seeking for Open-Domain Question Answering
TVR - Transformation driven Visual Reasoning
Project: MatchZoo - Open ToolKit for Deep Text Matching & Neural IR
Project: TextNet - Open Framework of Deep Learning for Text (C++)

Hornors and Awards

Academic Services

  • PC member for AAAI2019
  • PC member for ACML2019
  • PC member for CCIR2018
  • Reviewer for: TIST

2015 and before

MatchingZoo is a toolkit for text matching. It was developed with a focus on facilitate the designing, comparing and sharing of deep text matching models. The architecture of the MatchZoo toolit is depicited in Figure. There are three major modules in the toolkit, namely data preparation, model construction, training and evaluation, respectively. These three modules are actually organized as a pipeline of data flow.


Yixing Fan, Liang Pang, Jianpeng Hou, Jiafeng Guo, Yanyan Lan and Xueqi Cheng. MatchZoo: A Toolkit for Deep Text Matching. Neu-IR: The SIGIR 2017 Workshop on Neural Information Retrieval, Tokyo, 2017.

A platform for not only reporting the performances of state-of-the-art algorithms in different domains, but also providing the corresponding datasets, codes, and scripts that can make the experimental results reproducible.