KIT-TIP-NLP 在 MultiPride 上的持续学习：多语言基础模型
KIT-TIP-NLP at MultiPride: Continual Learning with Multilingual Foundation Model
AI总结：
本文提出了一种多阶段框架，用于检测社交媒体中多语言的重新使用侮辱性语言。该框架解决了跨英语、西班牙语和意大利语推文中识别重新使用与非重新使用LGBTQ+相关侮辱性语言的挑战，通过数据驱动的模型选择、语义保留的增强、归纳迁移学习和领域特定知识注入等方法，提高了多语言情感表达的识别能力。
摘要：
本文提出了一种多阶段框架，用于检测多语言社交媒体中重新使用的侮辱性语言。该框架解决了在英语、西班牙语和意大利语推文中识别重新使用与非重新使用LGBTQ+相关侮辱性语言的挑战。该框架处理了三个交织的方法学挑战：数据稀缺、类别不平衡和跨语言的情感表达差异。该框架整合了通过交叉验证的数据驱动模型选择、通过回译的语义保留增强、具有动态周期级欠采样的归纳迁移学习，以及通过掩码语言模型注入的领域特定知识。系统评估了八个多语言嵌入模型，XLM-RoBERTa被选为基础模型，基于宏平均F1分数。通过GPT-4o-mini回译进行的数据增强有效将训练语料库增加了三倍，同时保留了语义内容和类别分布比例。该框架生成了四个最终运行用于评估，其中RUN 1是带有增强和欠采样的归纳迁移学习，RUN 2是带有掩码语言模型预训练，RUN 3和RUN 4是通过语言特定决策阈值优化的先前预测。语言特定的阈值优化表明，最优决策边界在不同语言中存在显著差异。这反映了模型置信度分数的分布差异和重新使用语言使用的语言差异。基于阈值的优化在不需模型重新训练的情况下，带来了2-5%的绝对F1提升。该方法完全可复现，所有代码和实验设置可在https://github.com/rbg-research/MultiPRIDE-Evalita-2026上找到。
英文摘要：
This paper presents a multi-stage framework for detecting reclaimed slurs in multilingual social media discourse. It addresses the challenge of identifying reclamatory versus non-reclamatory usage of LGBTQ+-related slurs across English, Spanish, and Italian tweets. The framework handles three intertwined methodological challenges like data scarcity, class imbalance, and cross-linguistic variation in sentiment expression. It integrates data-driven model selection via cross-validation, semantic-preserving augmentation through back-translation, inductive transfer learning with dynamic epoch-level undersampling, and domain-specific knowledge injection via masked language modeling. Eight multilingual embedding models were evaluated systematically, with XLM-RoBERTa selected as the foundation model based on macro-averaged F1 score. Data augmentation via GPT-4o-mini back-translation to alternate languages effectively tripled the training corpus while preserving semantic content and class distribution ratios. The framework produces four final runs for the evaluation purposes where RUN 1 is inductive transfer learning with augmentation and undersampling, RUN 2 with masked language modeling pre-training, RUN 3 and RUN 4 are previous predictions refined via language-specific decision thresholds optimized via ROC analysis. Language-specific threshold refinement reveals that optimal decision boundaries vary significantly across languages. This reflects distributional differences in model confidence scores and linguistic variation in reclamatory language usage. The threshold-based optimization yields 2-5% absolute F1 improvement without requiring model retraining. The methodology is fully reproducible, with all code and experimental setup available at https://github.com/rbg-research/MultiPRIDE-Evalita-2026.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
