6月9日学术报告信息（Nianwen Xue，Brandeis University）
报告题目：Towards a universal meaning representation for natural language processing
报告人简介：Nianwen Xue is an Associate Professor in the Computer Science Department and the Language and Linguistics Program at Brandeis University. Dr. Xue directs the Chinese Language Processing Group in the Computer Science Deparment. Before joining Brandeis, Dr. Xue was a Research Assistant Professor in the Department of Linguistics and the Center for Computational Language and Education Research (CLEAR) at the University of Colorado at Boulder. Prior to that, he was a postdoctoral fellow in the Institute for Research in Cognitive Science and the Department of Computer and Information Science at the University of Pennsylvania. He received his PhD in Linguistics from University of Delaware. His research interests include syntactic, semantic, temporal and discourse annotation, semantic parsing, discourse analysis and Machine Translation. His research has been funded by NSF, DARPA, and IARPA. Dr. Xue is currently the Editor-in-Chief of TALLIP and serves on the editorial boards of LRE, and Lingua Sinica.
报告摘要：Understanding the meaning of natural language sentences by machine has long been one of the central goals in the field of natural language processing (NLP), and the development of such a technology would have a transformative effect on NLP specifically, and on Artificial Intelligence in general. With machine learning being the dominant approach in the field, the challenge has been to identify a meaning representation that can be used to consistently annotate a large amount of natural language data in multiple languages that can be used to train machine learning algorithms. In this talk, I will examine a few semantic annotation projects that have paved the way for a potential universal meaning representation that is scalable, learnable, cross-linguistically valid, and expressive enough to capture the major aspects of natural language semantics. I will start by discussing the predicate-argument structure annotation of the Proposition Bank (PropBank) and the Chinese Proposition Bank (CPB) projects. I will then discuss the AbstractMeaning Representation (AMR), a specification language that is largely based on the predicate-argument structure annotation in the Proposition Bank but is able to represent whole-sentence meaning with a single rooted, acyclic, directed graph. I argue that while the “graph” aspect of AMR has received much attention, it is the “abstract” nature of AMR that distinguishes itself from competing meaning representations (e.g., Universal Dependency, graph-based semantic dependency) and makes it a more plausible candidate based on which a universal meaning representation can be developed. Time permitting, I will also talk about our work in parsing AMRs and extending AMR to Chinese.
- 上一篇：6月1日学术报告信息（Bingsheng Zhang，Lancaster University）
- 下一篇：6月13日学术报告信息（Dr. Arun Kumar Sangaiah，VIT University, India）