Nile-Chat：面向阿拉伯文与拉丁文字体的埃及语语言模型
Nile-Chat: Egyptian Language Models for Arabic and Latin Scripts
AI总结：
该研究针对埃及方言提出Nile-Chat系列双字体LLM，通过Branch-Train-MiX策略构建MoE模型，在新基准上显著优于现有模型，为双字体语言适配提供全面方法并公开资源。
摘要：
我们引入了Nile-Chat-4B、3x4B-A6B和12B，这是一组针对埃及方言的大型语言模型（LLMs），其独特设计在于能够理解和生成以阿拉伯文与拉丁文字体书写的文本。具体而言，针对Nile-Chat-3x4B-A6B，我们提出了一种新颖的语言适配方法，即利用Branch-Train-MiX策略将字体专用专家模型合并为单一的混合专家（MoE）模型。我们的Nile-Chat模型在新引入的涵盖理解与生成任务的埃及语评估基准上，显著优于LLaMa、Jais和ALLaM等领先的多语言及阿拉伯语LLM。值得注意的是，我们的12B模型在拉丁文字体基准上比Qwen2.5-14B-Instruct高出14.4%的性能。所有资源均公开可用。我们认为这项工作为将LLM适配到双字体语言提供了全面方法，解决了现代LLM开发中常被忽视的方面。
英文摘要：
We introduce Nile-Chat-4B, 3x4B-A6B, and 12B, a collection of LLMs for Egyptian dialect, uniquely designed to understand and generate texts written in both Arabic and Latin scripts. Specifically, with Nile-Chat-3x4B-A6B, we introduce a novel language adaptation approach by leveraging the Branch-Train-MiX strategy to merge script-specialized experts, into a single MoE model. Our Nile-Chat models significantly outperform leading multilingual and Arabic LLMs, such as LLaMa, Jais, and ALLaM, on our newly introduced Egyptian evaluation benchmarks, which span both understanding and generative tasks. Notably, our 12B model yields a 14.4% performance gain over Qwen2.5-14B-Instruct on Latin-script benchmarks. All our resources are publicly available. We believe this work presents a comprehensive methodology for adapting LLMs to dual-script languages, addressing an often overlooked aspect in modern LLM development.
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。
