Handshakes AI Research at CASE 2021 Task 1: Exploring different approaches for multilingual tasks

TL;DR

This paper explores various approaches for multilingual socio-political and crisis event detection, emphasizing the benefits of leveraging multilingual data rather than treating languages separately.

Contribution

It demonstrates that embracing the multilingual nature of the tasks improves performance across all subtasks and languages.

Findings

Multilingual modeling enhances detection accuracy.

Unified training regimes outperform language-specific approaches.

The approach is validated across all subtasks and languages.

Abstract

The aim of the CASE 2021 Shared Task 1 (H\"urriyeto\u{g}lu et al., 2021) was to detect and classify socio-political and crisis event information at document, sentence, cross-sentence, and token levels in a multilingual setting, with each of these subtasks being evaluated separately in each test language. Our submission contained entries in all of the subtasks, and the scores obtained validated our research finding: That the multilingual aspect of the tasks should be embraced, so that modeling and training regimes use the multilingual nature of the tasks to their mutual benefit, rather than trying to tackle the different languages separately. Our code is available at https://github.com/HandshakesByDC/case2021/