A Joint Multitask Model for Morpho-Syntactic Parsing

TL;DR

This paper introduces a joint multitask model utilizing a shared encoder and specialized decoders for morpho-syntactic parsing, achieving top performance across nine languages in the UniDive 2025 shared task.

Contribution

It presents a novel multitask approach with a shared encoder and multiple decoders for morphological and syntactic analysis, improving performance on diverse languages.

Findings

Achieved highest overall scores on the shared task leaderboard.

Matching gold tokenization and content word identification are crucial.

Struggles with core grammatical cases and nominal features across languages.

Abstract

We present a joint multitask model for the UniDive 2025 Morpho-Syntactic Parsing shared task, where systems predict both morphological and syntactic analyses following novel UD annotation scheme. Our system uses a shared XLM-RoBERTa encoder with three specialized decoders for content word identification, dependency parsing, and morphosyntactic feature prediction. Our model achieves the best overall performance on the shared task's leaderboard covering nine typologically diverse languages, with an average MSLAS score of 78.7 percent, LAS of 80.1 percent, and Feats F1 of 90.3 percent. Our ablation studies show that matching the task's gold tokenization and content word identification are crucial to model performance. Error analysis reveals that our model struggles with core grammatical cases (particularly Nom-Acc) and nominal features across languages.