对习语语言的偏好是缓慢获得的——并且很快被遗忘：一项关于瑞典语的案例研究
Preferences for Idiomatic Language are Acquired Slowly -- and Forgotten Quickly: A Case Study on Swedish
AI总结：
本研究发现，语言模型在习语识别能力上发展缓慢且易遗忘，尤其在从英语适应到瑞典语时，机器翻译数据的微调会迅速削弱其对习语的偏好。
摘要：
在本研究中，我们探讨了语言模型在预训练和从英语适应到瑞典语过程中，如何发展对习语与语言上可接受的瑞典语之间的偏好。为此，我们从头开始训练瑞典语模型，并通过微调英语预训练模型，利用最小对进行探测，以评估它们在不同检查点的偏好。对于语言可接受性，我们将现有基准测试转换为最小对格式。为了评估习语性，我们引入了两个新的数据集：一个对比常规化习语与合理变体，另一个对比习语瑞典语与翻译语。我们的发现表明，习语能力的出现比其他语言能力更慢，包括语法和词汇正确性。尽管更长的训练对大多数任务产生边际效益，但与习语相关的性能继续改善，尤其是在测试的最大型（8B）中。然而，使用从英语机器翻译的数据进行指令微调——对于缺乏或没有本地指令数据的语言而言是常见方法——会导致模型迅速失去对习语语言的偏好。
英文摘要：
In this study, we investigate how language models develop preferences for \textit{idiomatic} as compared to \textit{linguistically acceptable} Swedish, both during pretraining and when adapting a model from English to Swedish. To do so, we train models on Swedish from scratch and by fine-tuning English-pretrained models, probing their preferences at various checkpoints using minimal pairs that differ in linguistic acceptability or idiomaticity. For linguistic acceptability, we adapt existing benchmarks into a minimal-pair format. To assess idiomaticity, we introduce two novel datasets: one contrasting conventionalized idioms with plausible variants, and another contrasting idiomatic Swedish with Translationese. Our findings suggest that idiomatic competence emerges more slowly than other linguistic abilities, including grammatical and lexical correctness. While longer training yields diminishing returns for most tasks, idiom-related performance continues to improve, particularly in the largest model tested (8B). However, instruction tuning on data machine-translated from English -- the common approach for languages with little or no native instruction data -- causes models to rapidly lose their preference for idiomatic language.
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。
