Multitask Parsing Across Semantic Representations

TL;DR

This paper demonstrates that multitask learning with auxiliary semantic parsing tasks improves UCCA parsing performance across multiple languages and settings, leveraging a unified transition-based system.

Contribution

It introduces a unified multitask learning approach that enhances semantic parsing performance across diverse representations and languages.

Findings

Multitask learning significantly improves UCCA parsing accuracy.

The approach works across three languages and various domain settings.

Unified transition-based system effectively handles multiple semantic representations.

Abstract

The ability to consolidate information of different types is at the core of intelligence, and has tremendous practical value in allowing learning for one task to benefit from generalizations learned for others. In this paper we tackle the challenging task of improving semantic parsing performance, taking UCCA parsing as a test case, and AMR, SDP and Universal Dependencies (UD) parsing as auxiliary tasks. We experiment on three languages, using a uniform transition-based system and learning architecture for all parsing tasks. Despite notable conceptual, formal and domain differences, we show that multitask learning significantly improves UCCA parsing in both in-domain and out-of-domain settings.