针对低资源场景下的极化检测难题，提出阈值与加权结合的优化方法。
Not All Pretraining are Created Equal: Threshold Tuning and Class Weighting for Imbalanced Polarization Tasks in Low-Resource Settings
- 采用多语言模型与标签加权损失函数，应对严重类别不平衡。
- 在英文二分类任务上达到0.8032的宏平均F1，多标签任务最高0.556。
- 适合关注社交媒体极化分析、低资源语言处理的研究者。
本文提交至SemEval-2025极化共享任务，针对社交媒体文本中的极化检测与分类问题，构建基于Transformer的英文与斯瓦希里语系统，涵盖三项子任务：二分类极化检测、多标签目标类型分类及多标签表现形式识别。方法融合多语言与非洲语言专用模型（mDeBERTa-v3-base、SwahBERT、AfriBERTa-large），使用类别加权损失函数、迭代分层数据划分及逐标签阈值调优，以缓解严重类别不平衡。最佳配置mDeBERTa-v3-base在二分类验证集上取得0.8032的宏平均F1，多标签任务表现优异（最高0.556宏平均F1）。错误分析显示，隐含极化、代码转换及政治激烈讨论与真实极化的区分仍是主要挑战。
原文摘要 · Abstract (English)
This paper describes my submission to the Polarization Shared Task at SemEval-2025, which addresses polarization detection and classification in social media text. I develop Transformer-based systems for English and Swahili across three subtasks: binary polarization detection, multi-label target type classification, and multi-label manifestation identification. The approach leverages multilingual and African language-specialized models (mDeBERTa-v3-base, SwahBERT, AfriBERTa-large), class-weighted loss functions, iterative stratified data splitting, and per-label threshold tuning to handle severe class imbalance. The best configuration, mDeBERTa-v3-base, achieves 0.8032 macro-F1 on validation for binary detection, with competitive performance on multi-label tasks (up to 0.556 macro-F1). Error analysis reveals persistent challenges with implicit polarization, code-switching, and distinguishing heated political discourse from genuine polarization.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。