arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2402.09759cs.CLcs.AI
收藏

高效的语言自适应预训练:扩展用于波兰语的最先进大型语言模型

Efficient Language Adaptive Pre-training: Extending State-of-the-Art Large Language Models for Polish

Apostroph Group(Apostroph集团)

Szymon Ruciński

AI总结:

本研究通过LAPT方法微调英语基础LLM以生成波兰语文本,训练的Curie-7B-v1模型在KLEJ基准上表现优异且资源消耗极低,证明仅训练1.2%参数即可高效为现有LLM添加新语言。

摘要:

本研究探讨了微调基础英语大型语言模型(LLMs)以生成波兰语文本的潜力。第一步涉及在3.11 GB的高质量数据集上进行语言自适应预训练(LAPT),该数据集包含2.76亿个波兰语token。LAPT之后进行了额外的微调,旨在解决九项KLEJ挑战。我们训练的模型Curie-7B-v1不仅以3.02的最低困惑度生成波兰语文本,在基于解码器的波兰语模型中表现最佳,而且其性能也紧逼最好的波兰语编码器-解码器模型,在9项任务中有8项的差距不到2%。Curie-7B-v1仅使用了典型数据集约2-3%的规模来学习波兰语。LAPT使用消费级GPU在不到五天内完成,突显了该方法的高效性。该模型的波兰语熟练度得到显著提升,证明了仅通过训练1.2%的参数即可向现有LLMs添加新语言的可行性。为促进社区的协作进展,该模型已作为开源发布。

英文摘要:

This study explores the potential of fine-tuning foundational English Large Language Models (LLMs) for generating Polish text. The first step involves Language Adaptive Pre-training (LAPT) on a high-quality dataset of 3.11 GB, consisting of 276 million Polish tokens. The LAPT is followed by additional fine-tuning aimed at solving nine KLEJ challenges. Our trained model Curie-7B-v1 not only generates Polish text with the lowest perplexity of 3.02 among decoder-based Polish models but also closely rivals the performance of the best Polish encoder-decoder models with a less than 2% gap on 8 out of 9 tasks. Curie-7B-v1 used approximately 2-3% of a typical dataset size to learn Polish. The LAPT was completed in less than five days using a consumer GPU, highlighting the method's efficiency. The proficiency of the model in Polish was significantly enhanced, demonstrating the viability of this approach for adding new languages to existing LLMs by training just 1.2% of its parameters. To contribute to the community's collaborative progress, the model has been released as open-source.

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。

↑