Enhanced Universal Dependency Parsing with Second-Order Inference and Mixture of Training Data

TL;DR

This paper introduces an enhanced graph-based dependency parser with second-order inference, leveraging mixed training data for low-resource languages, achieving state-of-the-art results after fixing submission issues.

Contribution

The paper proposes a novel dependency parsing system that combines second-order inference with mixed training data to improve performance on low-resource languages.

Findings

Significant improvement in Tamil dependency parsing performance.

Our system outperforms the top-ranked system after correction.

Second-order inference enhances parsing accuracy.

Abstract

This paper presents the system used in our submission to the \textit{IWPT 2020 Shared Task}. Our system is a graph-based parser with second-order inference. For the low-resource Tamil corpus, we specially mixed the training data of Tamil with other languages and significantly improved the performance of Tamil. Due to our misunderstanding of the submission requirements, we submitted graphs that are not connected, which makes our system only rank \textbf{6th} over 10 teams. However, after we fixed this problem, our system is 0.6 ELAS higher than the team that ranked \textbf{1st} in the official results.