Few-shot text-based emotion detection

TL;DR

This paper presents the Unibuc-NLP team's approach to few-shot text-based emotion detection using large language models, achieving competitive results across multiple languages in the SemEval 2025 Task 11.

Contribution

The paper explores the effectiveness of large language models with few-shot prompting and fine-tuning for emotion detection in multilingual text datasets.

Findings

Achieved an F1-macro of 0.7546 for English emotion detection

Lower performance on Portuguese and Emakhuwa subsets

Demonstrated the potential of large language models in multilingual emotion detection

Abstract

This paper describes the approach of the Unibuc - NLP team in tackling the SemEval 2025 Workshop, Task 11: Bridging the Gap in Text-Based Emotion Detection. We mainly focused on experiments using large language models (Gemini, Qwen, DeepSeek) with either few-shot prompting or fine-tuning. With our final system, for the multi-label emotion detection track (track A), we got an F1-macro of $0.7546$ (26/96 teams) for the English subset, $0.1727$ (35/36 teams) for the Portuguese (Mozambican) subset and $0.325$ (\textbf{1}/31 teams) for the Emakhuwa subset.