TY - GEN
T1 - Evaluation of Pretrained and Instruction-Based Pretrained Models for Emotion Detection in Arabic Social Media Text
AU - Rafiul Biswas, Md
AU - Ibrahim, Shimaa
AU - Bessghaier, Mabrouka
AU - Zaghoauni, Wajdi
N1 - Publisher Copyright:
© 2025 Incoma Ltd. All rights reserved.
PY - 2025/9/10
Y1 - 2025/9/10
N2 - This study evaluates three approaches—instruction prompting of large language models (LLMs), instruction fine-tuning of LLMs, and transformer-based pretrained models on emotion detection in Arabic social media text. We compare pretrained transformer models like AraBERT, CaMelBERT, and XLM-RoBERTa with instruction prompting with advanced LLMs like GPT-4o, Gemini, Deepseek, and Fanar, and instruction fine-tuning approaches with LLMs like Llama 3.1, Mistral, and Phi. With a highly preprocessed dataset of 10,000 labeled Arabic tweets with overlapping emotional labels, our findings reveal that transformer-based pretrained models outperform instruction prompting and instruction fine-tuning approaches. Instruction prompts leverage general linguistic skills with maximum efficiency but fall short in detecting subtle emotional contexts. Instruction fine-tuning is more specific but trails behind pretrained transformer models. Our findings establish the need for optimized instruction-based approaches and underscore the important role played by domain-specific transformer architectures in accurate Arabic emotion detection.
AB - This study evaluates three approaches—instruction prompting of large language models (LLMs), instruction fine-tuning of LLMs, and transformer-based pretrained models on emotion detection in Arabic social media text. We compare pretrained transformer models like AraBERT, CaMelBERT, and XLM-RoBERTa with instruction prompting with advanced LLMs like GPT-4o, Gemini, Deepseek, and Fanar, and instruction fine-tuning approaches with LLMs like Llama 3.1, Mistral, and Phi. With a highly preprocessed dataset of 10,000 labeled Arabic tweets with overlapping emotional labels, our findings reveal that transformer-based pretrained models outperform instruction prompting and instruction fine-tuning approaches. Instruction prompts leverage general linguistic skills with maximum efficiency but fall short in detecting subtle emotional contexts. Instruction fine-tuning is more specific but trails behind pretrained transformer models. Our findings establish the need for optimized instruction-based approaches and underscore the important role played by domain-specific transformer architectures in accurate Arabic emotion detection.
UR - https://www.scopus.com/pages/publications/105034194003
U2 - 10.26615/978-954-452-098-4-019
DO - 10.26615/978-954-452-098-4-019
M3 - Conference contribution
AN - SCOPUS:105034194003
T3 - International Conference Recent Advances in Natural Language Processing, RANLP
SP - 158
EP - 165
BT - Proceedings of the 15th International Conference on Recent Advances in Natural Language Processing - Natural Language Processing in the Generative AI Era, RANLP 2025
A2 - Angelova, Galia
A2 - Kunilovskaya, Maria
A2 - Escribe, Marie
A2 - Mitkov, Ruslan
PB - Incoma Ltd
T2 - 15th International Conference on Recent Advances in Natural Language Processing - Natural Language Processing in the Generative AI Era, RANLP 2025
Y2 - 8 September 2025 through 10 September 2025
ER -