基于语言学信息的参数聚合
Typologically Informed Parameter Aggregation
AI总结：
TIPA通过聚合现有适配器实现无需训练的跨语言迁移，有效提升低资源语言性能。
摘要：
大规模多语言语言模型能够实现跨语言泛化，但在低资源语言和未见语言上表现不佳。尽管基于适配器的微调提供了一种参数高效的方法，但大规模训练语言特定的适配器仍然成本高昂。我们引入了基于语言学信息的参数聚合（TIPA），这是一种无需训练的方法，通过按语言学相似性加权聚合现有适配器来构建代理语言适配器。将TIPA整合到MAD-X框架中，这些代理使零样本跨语言转移成为可能，而无需额外训练。我们在五个NLP任务和超过230种语言上评估了TIPA。TIPA在大多数情况下均优于或匹配英语微调或选择最接近的语言适配器。我们看到在缺乏专用适配器的语言上收益最大。我们的结果表明，基于语言学信息的聚合为无需训练的语言特定模块提供了一种可行的替代方案。
英文摘要：
Massively multilingual language models enable cross-lingual generalization but underperform on low-resource and unseen languages. While adapter-based fine-tuning offers a parameter-efficient solution, training language-specific adapters at scale remains costly. We introduce Typologically Informed Parameter Aggregation (TIPA), a training-free method that constructs proxy language adapters by aggregating existing ones, weighted by typological similarity. Integrated into the MAD-X framework, these proxies enable zero-shot cross-lingual transfer without additional training. We evaluate TIPA on five NLP tasks and over 230 languages. TIPA consistently outperforms or matches baselines such as English-only fine-tuning or selecting the typologically closest language adapter. We see the largest gains for languages lacking dedicated adapters. Our results demonstrate that typologically informed aggregation provides a viable alternative to language-specific modules without any training needed.
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。
