A list of Abstract meaning representation (AMR) resources: research papers, code, data, applications, etc. The list is not limited to Abstract meaning representation exclusively. It also includes work highly related to AMR, such as translation and information extraction.
抽象意义表示(Abstract Meaning Representation,AMR)是一种将句子的意义(时间地点谁对谁怎样地做了什么)表示为以概念为节点的单源有向无环图的语言学框架。AMR正在引起学术界越来越广泛的关注,已经涌现了许多利用AMR进行机器翻译、QA、关系提取等应用的工作。
Suppose we have the following input sentence:
The boy wants the girl to believe him.
AMR aims to make the following format:
(w / want-01
:ARG0 (b / boy)
:ARG1 (b2 / believe-01
:ARG0 (g / girl)
:ARG1 b))
-
"Abstract Meaning Representation for Sembanking" - ACL 2013
Laura Banarescu, Claire Bonial, Shu Cai
smatch metric is developed for comparing one AMR against another. It computes an f-score over the semantic triples inside AMRs.
-
"Smatch: an Evaluation Metric for Semantic Feature Structures," ACL 2013 (smatch code ,demo)
S. Cai and K. Knight
-
"Transition-Based Dependency Parsing with Stack Long Short-Term Memory," ACL 2015 code
Chris Dyer, Miguel Ballesteros, Wang Ling, Austin Matthews, Noah A. Smith
-
"CAMR at SemEval-2016 Task 8: An Extended Transition-based AMR Parser," ACL 2016
Chuan Wang, Sameer Pradhan, Xiaoman Pan, Heng Ji, Nianwen Xue
-
"Transition-based Parsing with Stack-Transformers," ACL 2020
Ramón Fernandez Astudillo, Miguel Ballesteros, Tahira Naseem, Austin Blodgett, Radu Florian
-
"Deep biaffine attention for neural dependency parsing", ACL 2017 codecode
Timothy Dozat, Christopher D. Manning
-
"AMR Parsing via Graph-Sequence Iterative Inference," ACL 2020 code
Deng Cai, Wai Lam
-
"Levi Graph AMR Parser using Heterogeneous Attention", ACL 2021 code
Han He, Jinho D. Choi
AMR graph output has been shown to be a useful input for many downstream tasks. In this section, several downstream tasks that benefited from AMR output are listed.
-
AMR 2.0 (LDC2017T10)
-
AMR 3.0(LDC2020T02)
-
CAMR 2.0 由布兰迪斯大学和南京师范大学联合标注的中文抽象意义表示语料库2.0