Arabic Stable LM：使 Stable LM 2 1.6B 适配阿拉伯语
Arabic Stable LM: Adapting Stable LM 2 1.6B to Arabic
AI总结：
本文提出基座与聊天两版 Arabic Stable LM 1.6B，以轻量阿拉伯语中心 LLM 降低硬件需求和推理延迟，并验证合成对话指令微调可使其击败最高 8 倍参数的模型。
摘要：
大语言模型（LLM）已在自然语言处理（NLP）的多个领域展现出令人瞩目的成果，但主要集中于英语。近来，越来越多的 LLM 纳入了更大比例的多语言文本，以表示低资源语言。在阿拉伯语 NLP 中，过去两年已有若干以阿拉伯语为中心的 LLM 在多个基准测试上取得显著成果。然而，与较小的 LLM 相比，大多数阿拉伯语 LLM 的参数量超过 70 亿，这提高了它们的硬件需求和推理延迟。本文介绍了 Arabic Stable LM 1.6B，包含基座版和聊天版，是一个小巧但强大的以阿拉伯语为中心的 LLM。我们的 Arabic Stable LM 1.6B 聊天模型在多个基准测试上取得令人瞩目的成果，击败了参数量最高达其 8 倍的多个模型。此外，我们通过用一个大型合成对话数据集增强微调数据，展示了混入合成指令微调数据的益处。
英文摘要：
Large Language Models (LLMs) have shown impressive results in multiple domains of natural language processing (NLP) but are mainly focused on the English language. Recently, more LLMs have incorporated a larger proportion of multilingual text to represent low-resource languages. In Arabic NLP, several Arabic-centric LLMs have shown remarkable results on multiple benchmarks in the past two years. However, most Arabic LLMs have more than 7 billion parameters, which increases their hardware requirements and inference latency, when compared to smaller LLMs. This paper introduces Arabic Stable LM 1.6B in a base and chat version as a small but powerful Arabic-centric LLM. Our Arabic Stable LM 1.6B chat model achieves impressive results on several benchmarks beating multiple models with up to 8x the parameters. In addition, we show the benefit of mixing in synthetic instruction tuning data by augmenting our fine-tuning data with a large synthetic dialogue dataset.
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。
