LGCLMLMay 27, 2019

Sequential Graph Dependency Parser

arXiv:1905.10930v2997 citations
Originality Incremental advance
AI Analysis

This addresses parsing challenges in NLP for languages with complex syntactic structures, but it is incremental as it blends existing parsing approaches.

The paper tackles non-projective dependency parsing by predicting edges incrementally without a pre-specified order, achieving near state-of-the-art results on both projective and non-projective languages.

We propose a method for non-projective dependency parsing by incrementally predicting a set of edges. Since the edges do not have a pre-specified order, we propose a set-based learning method. Our method blends graph, transition, and easy-first parsing, including a prior state of the parser as a special case. The proposed transition-based method successfully parses near the state of the art on both projective and non-projective languages, without assuming a certain parsing order.

Foundations

The foundational work for this paper's niche, ranked by how specifically the neighbourhood builds on it — not by global fame.

Your Notes