{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/pali-nlp-at-semeval-2022-task-4","title":"PALI-NLP at SemEval-2022 Task 4: Discriminative Fine-tuning of Transformers for Patronizing and Condescending Language Detection","arxiv_id":"2203.04616","date":"2022-03-09","proceeding":"SemEval (NAACL) 2022 7","authors":["Dou Hu","Mengyuan Zhou","Xiyang Du","Mengfei Yuan","Meizhi Jin","Lianxin Jiang","Yang Mo","Xiaofeng Shi"],"abstract":"Patronizing and condescending language (PCL) has a large harmful impact and is difficult to detect, both for human judges and existing NLP systems. At SemEval-2022 Task 4, we propose a novel Transformer-based model and its ensembles to accurately understand such language context for PCL detection. To facilitate comprehension of the subtle and subjective nature of PCL, two fine-tuning strategies are applied to capture discriminative features from diverse linguistic behaviour and categorical distribution. The system achieves remarkable results on the official ranking, including 1st in Subtask 1 and 5th in Subtask 2. Extensive experiments on the task demonstrate the effectiveness of our system and its strategies.","url_abs":"https://arxiv.org/abs/2203.04616v2","url_pdf":"https://arxiv.org/pdf/2203.04616v2.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[],"tasks":[{"task_slug":"binary-condescension-detection","task_name":"Binary Condescension Detection"},{"task_slug":"multi-label-condescension-detection","task_name":"Multi-label Condescension Detection"},{"task_slug":"semeval-2022-task-4-1-binary-pcl-detection","task_name":"SemEval-2022 Task 4-1 (Binary PCL Detection)"},{"task_slug":"semeval-2022-task-4-2-multi-label-pcl","task_name":"SemEval-2022 Task 4-2 (Multi-label PCL Detection)"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/binary-condescension-detection-on-dpm","task":"Binary Condescension Detection","dataset":"DPM","model":"BERT-PCL","rank_in_archive_order":1,"of":5,"metrics":{"F1-score":"63.69"},"uses_additional_data":false},{"leaderboard":"/sota/multi-label-condescension-detection-on-dpm","task":"Multi-label Condescension Detection","dataset":"DPM","model":"BERT-PCL","rank_in_archive_order":2,"of":5,"metrics":{"Macro-F1":"43.28"},"uses_additional_data":false}],"syntology":{"syntology_url":"https://syntology.ai/paper/2203.04616","atlas_url":"https://app.syntology.ai/?focus=2203.04616","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}