MKJ在SemEval-2026任务9中的表现：多语言极化检测中通用、专业和集成策略的比较研究
MKJ at SemEval-2026 Task 9: A Comparative Study of Generalist, Specialist, and Ensemble Strategies for Multilingual Polarization
AI总结：
本文比较了多语言极化检测中通用、专业和集成策略的效果，探讨了多语言通用模型与语言特定专家模型的差异，以及在不同语言任务中的表现。
摘要：
我们对SemEval-2026任务9（子任务1）的22种语言多语言极化检测进行了系统研究，对比了多语言通用模型与语言特定专家模型以及混合集成模型。虽然标准通用模型如XLM-RoBERTa在 tokenizer 与目标文本匹配时表现良好，但在具有不同书写系统（如高棉语、奥里亚语）的情况下，单语言专家模型能取得显著优势。我们采用语言自适应框架，根据开发性能在多语言通用模型、语言特定专家模型和混合集成模型之间切换。此外，通过NLLB-200进行跨语言增强效果混合，通常表现不佳，不如原生架构选择，并在形态丰富的任务中有所下降。我们的最终系统在所有22个任务上实现了总体宏平均F1得分为0.796，平均准确率为0.826。代码和最终测试预测可在https://github.com/Maziarkiani/SemEval2026-Task9-Subtask1-Polarization上公开获取。
英文摘要：
We present a systematic study of multilingual polarization detection across 22 languages for SemEval-2026 Task 9 (Subtask 1), contrasting multilingual generalists with language-specific specialists and hybrid ensembles. While a standard generalist like XLM-RoBERTa suffices when its tokenizer aligns with the target text, it may struggle with distinct scripts (e.g., Khmer, Odia) where monolingual specialists yield significant gains. Rather than enforcing a single universal architecture, we adopt a language-adaptive framework that switches between multilingual generalists, language-specific specialists, and hybrid ensembles based on development performance. Additionally, cross-lingual augmentation via NLLB-200 yielded mixed results, often underperforming native architecture selection and degrading morphologically rich tracks. Our final system achieves an overall macro-averaged F1 score of 0.796 and an average accuracy of 0.826 across all 22 tracks. Code and final test predictions are publicly available at: https://github.com/Maziarkiani/SemEval2026-Task9-Subtask1-Polarization.
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。
