{"ID":2837933,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2511.18324","arxiv_id":"2511.18324","title":"Gradient Masters at BLP-2025 Task 1: Advancing Low-Resource NLP for Bengali using Ensemble-Based Adversarial Training for Hate Speech Detection","abstract":"This paper introduces the approach of \"Gradient Masters\" for BLP-2025 Task 1: \"Bangla Multitask Hate Speech Identification Shared Task\". We present an ensemble-based fine-tuning strategy for addressing subtasks 1A (hate-type classification) and 1B (target group classification) in YouTube comments. We propose a hybrid approach on a Bangla Language Model, which outperformed the baseline models and secured the 6th position in subtask 1A with a micro F1 score of 73.23% and the third position in subtask 1B with 73.28%. We conducted extensive experiments that evaluated the robustness of the model throughout the development and evaluation phases, including comparisons with other Language Model variants, to measure generalization in low-resource Bangla hate speech scenarios and data set coverage. In addition, we provide a detailed analysis of our findings, exploring misclassification patterns in the detection of hate speech.","short_abstract":"This paper introduces the approach of \"Gradient Masters\" for BLP-2025 Task 1: \"Bangla Multitask Hate Speech Identification Shared Task\". We present an ensemble-based fine-tuning strategy for addressing subtasks 1A (hate-type classification) and 1B (target group classification) in YouTube comments. We propose a hybrid a...","url_abs":"https://arxiv.org/abs/2511.18324","url_pdf":"https://arxiv.org/pdf/2511.18324v1","authors":"[\"Syed Mohaiminul Hoque\",\"Naimur Rahman\",\"Md Sakhawat Hossain\"]","published":"2025-11-23T07:29:09Z","proceeding":"cs.CL","tasks":"[\"cs.CL\"]","methods":"[\"Language Model\"]","has_code":false}