Lingo_Research_Group at SemEval-2026 Task 9: Evaluating Prompt Variants for Polarization Detection
For researchers in multilingual NLP and social media analysis, this work provides a systematic comparison of prompt designs for polarization detection, but the results are incremental and task-specific.
The paper evaluates twelve prompt variants for multilingual polarization detection using aya-101 and Gemma3-27B, achieving average macro F1-scores of 0.762, 0.587, and 0.444 on three subtasks across 22 languages. Prompt-based approaches work well for coarse-grained detection but struggle with fine-grained and multi-label classification.
Our submission presented in this paper is for SemEval-2026 Task 9: Multilingual Text Classification Challenge - Polarization Detection and it covers all three subtasks: (1) binary polarization detection, (2) polarization type classification and (3) polarization manifestation identification. We adopt a systematic approach of research on short designed prompts by considering twelve designed prompts that are different in terminology clarity, detail of the definition, guidance of reasoning and in-context examples use. The experiments are conducted using aya-101 and Gemma3-27B, with the latter chosen for the submission at the end of the development through performance considerations. Our system has an average macro level F1-score of 0.762 on Subtask 1, 0.587 on Subtask 2 and 0.444 on Subtask 3 with the average accuracy of 0.819, 0.678 and 0.498, respectively, on the official test set averaged among 22 languages, respectively. With cross-task and cross-lingual analysis, we demonstrate that prompt-based approaches can be used effectively to detect coarse grained polarization but encounter more and more difficulties as far as fine-grained and multi-label sociolinguistic classification is concerned.