Addressing the Data Sparsity Issue in Neural AMR Parsing
This work addresses the data sparsity problem for researchers and practitioners in NLP, specifically in AMR parsing, and is incremental as it builds on existing neural attention models.
The paper tackles the data sparsity issue in neural AMR parsing by proposing a sequence-to-sequence model with methods to address this problem, achieving significant improvement over a baseline neural attention model and competitive results against state-of-the-art systems without extra linguistic resources.
Neural attention models have achieved great success in different NLP tasks. How- ever, they have not fulfilled their promise on the AMR parsing task due to the data sparsity issue. In this paper, we de- scribe a sequence-to-sequence model for AMR parsing and present different ways to tackle the data sparsity problem. We show that our methods achieve significant improvement over a baseline neural atten- tion model and our results are also compet- itive against state-of-the-art systems that do not use extra linguistic resources.