Gpachov at CheckThat! 2023: A Diverse Multi-Approach Ensemble for Subjectivity Detection in News Articles

TL;DR

This paper describes a multi-approach ensemble system for subjectivity detection in news articles, combining sentence embeddings, few-shot learning, and multilingual transformers to achieve high accuracy and competitive results.

Contribution

The paper introduces a diverse ensemble method that integrates three different models for subjectivity detection, demonstrating improved performance over individual approaches.

Findings

Achieved 0.77 macro F1 score on the test set.

Secured 2nd place in the English subtask.

Showed effectiveness of combining diverse models for NLP tasks.

Abstract

The wide-spread use of social networks has given rise to subjective, misleading, and even false information on the Internet. Thus, subjectivity detection can play an important role in ensuring the objectiveness and the quality of a piece of information. This paper presents the solution built by the Gpachov team for the CLEF-2023 CheckThat! lab Task~2 on subjectivity detection. Three different research directions are explored. The first one is based on fine-tuning a sentence embeddings encoder model and dimensionality reduction. The second one explores a sample-efficient few-shot learning model. The third one evaluates fine-tuning a multilingual transformer on an altered dataset, using data from multiple languages. Finally, the three approaches are combined in a simple majority voting ensemble, resulting in 0.77 macro F1 on the test set and achieving 2nd place on the English subtask.