Please Mind the Root: Decoding Arborescences for Dependency Parsing

TL;DR

This paper highlights the importance of the root constraint in dependency parsing, analyzes parser outputs across languages, and introduces an efficient algorithm to enforce the root constraint without increasing decoding complexity.

Contribution

It adapts a classical algorithm to ensure the root constraint in dependency parsing efficiently, improving the reliability of parser outputs.

Findings

State-of-the-art parsers often violate the root constraint, especially with smaller training sets.

Existing methods to enforce the constraint are inefficient, adding computational overhead.

The proposed algorithm enforces the root constraint without increasing decoding runtime.

Abstract

The connection between dependency trees and spanning trees is exploited by the NLP community to train and to decode graph-based dependency parsers. However, the NLP literature has missed an important difference between the two structures: only one edge may emanate from the root in a dependency tree. We analyzed the output of state-of-the-art parsers on many languages from the Universal Dependency Treebank: although these parsers are often able to learn that trees which violate the constraint should be assigned lower probabilities, their ability to do so unsurprisingly de-grades as the size of the training set decreases. In fact, the worst constraint-violation rate we observe is 24%. Prior work has proposed an inefficient algorithm to enforce the constraint, which adds a factor of n to the decoding runtime. We adapt an algorithm due to Gabow and Tarjan (1984) to dependency parsing, which satisfies the constraint without compromising the original runtime.