A Higher-Order Semantic Dependency Parser
This work addresses a computational bottleneck in semantic parsing for NLP researchers, offering an incremental improvement over existing methods.
The paper tackled the NP-hard problem of modeling higher-order features in semantic dependency parsing by using graph neural networks (GNNs) to aggregate information through multiple layers, resulting in a model that outperformed the previous state-of-the-art on the SemEval 2015 Task 18 English datasets.
Higher-order features bring significant accuracy gains in semantic dependency parsing. However, modeling higher-order features with exact inference is NP-hard. Graph neural networks (GNNs) have been demonstrated to be an effective tool for solving NP-hard problems with approximate inference in many graph learning tasks. Inspired by the success of GNNs, we investigate building a higher-order semantic dependency parser by applying GNNs. Instead of explicitly extracting higher-order features from intermediate parsing graphs, GNNs aggregate higher-order information concisely by stacking multiple GNN layers. Experimental results show that our model outperforms the previous state-of-the-art parser on the SemEval 2015 Task 18 English datasets.