# Unsupervised Dependency Parsing: Let's Use Supervised Parsers

arXiv: https://arxiv.org/abs/1504.04666 · HTML: https://mlsanity.com/paper/1504.04666/

## Key facts

- Cited by 3 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 18 Apr 2015.

## Built on by (most cited first, 3 of 3)

- [Semantic Role Labeling as Dependency Parsing: Exploring Latent Tree Structures Inside Arguments](https://mlsanity.com/paper/2110.06865/) (2021)
- [Not All Linearizations Are Equally Data-Hungry in Sequence Labeling Parsing](https://mlsanity.com/paper/2108.07556/) (2021)
- [Understanding Syntactic Generalization in Structure-inducing Language Models](https://mlsanity.com/paper/2508.07969/) (2025)

## Abstract (arXiv)

We present a self-training approach to unsupervised dependency parsing that reuses existing supervised and unsupervised parsing algorithms. Our approach, called `iterated reranking' (IR), starts with dependency trees generated by an unsupervised parser, and iteratively improves these trees using the richer probability models used in supervised parsing that are in turn trained on these trees. Our system achieves 1.8% accuracy higher than the state-of-the-part parser of Spitkovsky et al. (2013) on the WSJ corpus.