UoB-NLP at SemEval-2025 Task 11: Leveraging Adapters for Multilingual and Cross-Lingual Emotion Detection

TL;DR

The paper tackles multilingual and cross-lingual emotion detection by leveraging adapters with multilingual PLMs to achieve parameter-efficient fine-tuning. It introduces target-language-ready task adapters and language-family-based adapter strategies and demonstrates strong performance on low-resource African languages, with competitive results across 16 languages. The approach frequently outperforms large language models under few-shot settings while using far fewer parameters, underscoring the practicality of modular adapters for multilingual NLP. The work highlights the potential for scalable cross-lingual transfer and suggests future exploration of prompt-tuning with LLMs.

Abstract

Emotion detection in natural language processing is a challenging task due to the complexity of human emotions and linguistic diversity. While significant progress has been made in high-resource languages, emotion detection in low-resource languages remains underexplored. In this work, we address multilingual and cross-lingual emotion detection by leveraging adapter-based fine-tuning with multilingual pre-trained language models. Adapters introduce a small number of trainable parameters while keeping the pre-trained model weights fixed, offering a parameter-efficient approach to adaptation. We experiment with different adapter tuning strategies, including task-only adapters, target-language-ready task adapters, and language-family-based adapters. Our results show that target-language-ready task adapters achieve the best overall performance, particularly for low-resource African languages with our team ranking 7th for Tigrinya, and 8th for Kinyarwanda in Track A. In Track C, our system ranked 3rd for Amharic, and 4th for Oromo, Tigrinya, Kinyarwanda, Hausa, and Igbo. Our approach outperforms large language models in 11 languages and matches their performance in four others, despite our models having significantly fewer parameters. Furthermore, we find that adapter-based models retain cross-linguistic transfer capabilities while requiring fewer computational resources compared to full fine-tuning for each language.

Figures (2)

• Figure 1: Languages used in this study and their partitioning into language families for adapter training.
• Figure 2: Illustration of a multilingual task adapter (TA) with a target language-ready (TLR) module at a single PLM layer. The diagram shows language adapters (LAs) for K target languages alongside the emotion detection TA. This method is from Parovic2023.