低資源語言文本去毒化的Tatoxa系統：以韃靼語為例

摘要

文本去毒化，即对攻击性和有害内容进行自动检测与缓解，对于保障在线社区安全及保护用户至关重要。然而，鞑靼语等低资源语言在此领域鲜少受到研究关注。本文提出Tatoxa——一种面向鞑靼语文本去毒化的前沿系统，其性能在关键质量指标上超越了现有开源及专有商用大型语言模型。我们还引入了一个新的鞑靼语文本去毒化数据集，专为低资源环境下的微调与评估而设计。最后，跨语言迁移实验表明，即便拥有大规模的俄语语料库，从包括文化相近的俄语在内的其他语言进行迁移学习，其效果也显著劣于基于原生鞑靼语数据的训练。

English

Text detoxification, the automated detection and mitigation of abusive and harmful content, is essential for ensuring the safety of online communities and protecting users. However, low resource languages such as Tatar have received little research attention. In this paper we present Tatoxa, a novel state-of-the-art system for text detoxification in the Tatar language. Comparative experiments show that the proposed approach outperforms existing open source and proprietary commercial LLMs on key quality metrics. We also introduce a new dataset for text detoxification in Tatar, designed for fine tuning and evaluation in low resource settings. Finally, cross lingual transfer experiments indicate that transfer from other languages, including the culturally close Russian, performs significantly worse than training on native Tatar data even when a large Russian corpus is available.