Header

UZH-Logo

Maintenance Infos

CL-UZH at SemEval-2023 Task 10: Sexism Detection through Incremental Fine-Tuning and Multi-Task Learning with Label Descriptions


Goldzycher, Janis (2023). CL-UZH at SemEval-2023 Task 10: Sexism Detection through Incremental Fine-Tuning and Multi-Task Learning with Label Descriptions. In: Proceedings of the 17th International Workshop on Semantic Evaluation (SemEval-2023), Toronto, Canada, 13 July 2023 - 14 July 2023. Association for Computational Linguistics, 1562-1572.

Abstract

The widespread popularity of social media has led to an increase in hateful, abusive, and sexist language, motivating methods for the automatic detection of such phenomena. The goal of the SemEval shared task Towards Explainable Detection of Online Sexism (EDOS 2023) is to detect sexism in English social media posts (subtask A), and to categorize such posts into four coarse-grained sexism categories (subtask B), and eleven fine-grained subcategories (subtask C). In this paper, we present our submitted systems for all three subtasks, based on a multi-task model that has been fine-tuned on a range of related tasks and datasets before being fine-tuned on the specific EDOS subtasks. We implement multi-task learning by formulating each task as binary pairwise text classification, where the dataset and label descriptions are given along with the input text. The results show clear improvements over a fine-tuned DeBERTa-V3 serving as a baseline leading to F1-scores of 85.9% in subtask A (rank 13/84), 64.8% in subtask B (rank 19/69), and 44.9% in subtask C (26/63).

Abstract

The widespread popularity of social media has led to an increase in hateful, abusive, and sexist language, motivating methods for the automatic detection of such phenomena. The goal of the SemEval shared task Towards Explainable Detection of Online Sexism (EDOS 2023) is to detect sexism in English social media posts (subtask A), and to categorize such posts into four coarse-grained sexism categories (subtask B), and eleven fine-grained subcategories (subtask C). In this paper, we present our submitted systems for all three subtasks, based on a multi-task model that has been fine-tuned on a range of related tasks and datasets before being fine-tuned on the specific EDOS subtasks. We implement multi-task learning by formulating each task as binary pairwise text classification, where the dataset and label descriptions are given along with the input text. The results show clear improvements over a fine-tuned DeBERTa-V3 serving as a baseline leading to F1-scores of 85.9% in subtask A (rank 13/84), 64.8% in subtask B (rank 19/69), and 44.9% in subtask C (26/63).

Statistics

Citations

Dimensions.ai Metrics

Altmetrics

Downloads

0 downloads since deposited on 11 Dec 2023
0 downloads since 12 months

Additional indexing

Item Type:Conference or Workshop Item (Paper), original work
Communities & Collections:06 Faculty of Arts > Institute of Computational Linguistics
Dewey Decimal Classification:410 Linguistics
000 Computer science, knowledge & systems
Scopus Subject Areas:Physical Sciences > Computational Theory and Mathematics
Physical Sciences > Computer Science Applications
Physical Sciences > Theoretical Computer Science
Language:English
Event End Date:14 July 2023
Deposited On:11 Dec 2023 11:42
Last Modified:12 Dec 2023 21:00
Publisher:Association for Computational Linguistics
OA Status:Closed
Publisher DOI:https://doi.org/10.18653/v1/2023.semeval-1.216
Official URL:https://aclanthology.org/2023.semeval-1.216