Please Mind the Root: Decoding Arborescences for Dependency Parsing

by   Ran Zmigrod, et al.

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.


page 1

page 2

page 3

page 4


On Finding the K-best Non-projective Dependency Trees

The connection between the maximum spanning tree in a directed graph and...

Efficient Sampling of Dependency Structures

Probabilistic distributions over spanning trees in directed graphs are a...

Unbiased and Efficient Sampling of Dependency Trees

Distributions over spanning trees are the most common way of computation...

Is It Really Useful to Jointly Parse Constituency and Dependency Trees? A Revisit

This work visits the topic of jointly parsing constituency and dependenc...

Maximum Spanning Trees Are Invariant to Temperature Scaling in Graph-based Dependency Parsing

Modern graph-based syntactic dependency parsers operate by predicting, f...

Probing for Labeled Dependency Trees

Probing has become an important tool for analyzing representations in Na...

Minor changes make a difference: a case study on the consistency of UD-based dependency parsers

Many downstream applications are using dependency trees, and are thus re...

Please sign up or login with your details

Forgot password? Click here to reset