跨语言迁移效果受任务与数据设置影响，语言相似性并非决定因素。
Analyzing the Effect of Linguistic Similarity on Cross-Lingual Transfer: Tasks and Experimental Setups Matter
- 分析263种语言在三种NLP任务上的跨语言迁移表现
- 发现语言相似性对性能的影响因任务和输入表示而异
- 挑战“越相似语言越易迁移”的普遍认知，适合多语言研究者
跨语言迁移是低资源场景下扩充NLP训练数据的常用方法。然而，如何选择最优的跨语言数据仍不明确。以往研究多限于少数语言家族中的少量语言和单一任务，其结论是否可推广至更广泛的语言和任务尚不清楚。本文分析了来自多种语言家族的263种语言在三种主流NLP任务（词性标注、依存句法分析、主题分类）上的跨语言迁移表现。结果表明，语言相似性对迁移性能的影响受多种因素调节：具体任务类型、输入表示方式（单语或双语），以及语言相似性的定义方式。该研究揭示了跨语言迁移的复杂性，强调需根据具体场景评估数据选择策略。
原文摘要 · Abstract (English)
Cross-lingual transfer is a popular approach to increase the amount of training data for NLP tasks in a low-resource context. However, the best strategy to decide which cross-lingual data to include is unclear. Prior research often focuses on a small set of languages from a few language families and/or a single task. It is still an open question how these findings extend to a wider variety of languages and tasks. In this work, we analyze cross-lingual transfer for 263 languages from a wide variety of language families. Moreover, we include three popular NLP tasks: POS tagging, dependency parsing, and topic classification. Our findings indicate that the effect of linguistic similarity on transfer performance depends on a range of factors: the NLP task, the (mono- or multilingual) input representations, and the definition of linguistic similarity.
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。