Simpler but More Accurate Semantic Dependency Parsing

2018/07/03 by Timothy Dozat, Christopher D. Manning, Dozat, Timothy +1 · 4 citations
Computer Science · #Computation and Language (cs.CL) #FOS: Computer and information sciences #Natural Language Processing Techniques #Text Readability and Simplification #Topic Modeling

paper · pdf · doi:10.48550/arxiv.1807.01396

openalex publication_date 2018/07/03 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28

Abstract

While syntactic dependency annotations concentrate on the surface or functional structure of a sentence, semantic dependency annotations aim to capture between-word relationships that are more closely related to the meaning of a sentence, using graph-structured representations. We extend the LSTM-based syntactic parser of Dozat and Manning (2017) to train on and generate these graph structures. The resulting system on its own achieves state-of-the-art performance, beating the previous, substantially more complex state-of-the-art system by 0.6% labeled F1. Adding linguistically richer input representations pushes the margin even higher, allowing us to beat it by 1.9% labeled F1.

Citations

• Universal Semantic Parsing
• Dropout as a Bayesian Approximation: Representing Model Uncertainty in Deep Learning

Cited by

• End-to-End Argument Mining through Autoregressive Argumentative Structure Prediction
• Hierarchical Bracketing Encodings Work for Dependency Graphs
• ÚFAL at MRP 2020: Permutation-invariant Semantic Parsing in PERIN
• Dependency Parsing is More Parameter-Efficient with Normalization

Related

• Deep Multitask Learning for Semantic Dependency Parsing
• A Survey of Syntactic-Semantic Parsing Based on Constituent and Dependency Structures
• Parsing All: Syntax and Semantics, Dependencies and Spans
• Exploiting Rich Syntactic Information for Semantic Parsing with Graph-to-Sequence Model
• Graph-Based Decoding for Task Oriented Semantic Parsing