arXiv 2406.017712024-06-03cs.CL

超越英语的LLMs：利用跨语言反馈扩展LLMs的多语言能力

LLMs Beyond English: Scaling the Multilingual Capability of LLMs with Cross-Lingual Feedback

School of Computation, Information and Technology, TUM（慕尼黑工业大学计算、信息与技术学院）；Center for Information and Language Processing, LMU Munich（慕尼黑大学信息与语言处理中心）；Munich Center for Machine Learning（慕尼黑机器学习中心）；Bosch Center for Artificial Intelligence（博世人工智能中心）

Wen Lai, Mohsen Mesgar, Alexander Fraser

备注：: Accepted to Findings of ACL 2024. The code, datasets, and models are publicly available at https://github.com/boschresearch/ACL24-MLLM

阅读 PDF

arXiv 页面 ↗HTML 原文 ↗

AI总结：

本文提出xLLMs-100（xLLaMA-100和xBLOOM-100），通过构建100语言指令数据集和30语言跨语言反馈数据集，结合指令微调与DPO算法，将LLaMA和BLOOM扩展至100种语言，在五个基准上显著超越同类模型。

摘要：

为了将大型语言模型（LLMs）普及到大多数自然语言，必须使这些模型能够理解和生成多种语言的文本，特别是低资源语言。尽管最近的多语言LLMs在此类能力上表现出色，但由于缺乏低资源语言的训练数据，这些LLMs仍只支持有限数量的人类语言。此外，这些LLMs尚未与下游任务的人类偏好对齐，而这对于LLMs在英语中的成功至关重要。在本文中，我们介绍了xLLaMA-100和xBLOOM-100（统称xLLMs-100），将LLaMA和BLOOM的多语言能力扩展到100种语言。为此，我们构建了两个数据集：一个包含100种语言的多语言指令数据集，代表了迄今为止最大的语言覆盖范围；以及一个涵盖30种语言的跨语言人类反馈数据集。我们在构建的指令数据上执行多语言指令微调，并进一步在跨语言人类反馈数据集上使用DPO算法将LLMs与人类反馈对齐。我们在五个多语言基准上评估了xLLMs-100的多语言理解和生成能力。实验结果表明，xLLMs-100在各项基准上以显著优势持续超越同类模型，定义了支持100种语言的最新最先进多语言LLM。

英文摘要：

To democratize large language models (LLMs) to most natural languages, it is imperative to make these models capable of understanding and generating texts in many languages, in particular low-resource ones. While recent multilingual LLMs demonstrate remarkable performance in such capabilities, these LLMs still support a limited number of human languages due to the lack of training data for low-resource languages. Moreover, these LLMs are not yet aligned with human preference for downstream tasks, which is crucial for the success of LLMs in English. In this paper, we introduce xLLaMA-100 and xBLOOM-100 (collectively xLLMs-100), which scale the multilingual capabilities of LLaMA and BLOOM to 100 languages. To do so, we construct two datasets: a multilingual instruction dataset including 100 languages, which represents the largest language coverage to date, and a cross-lingual human feedback dataset encompassing 30 languages. We perform multilingual instruction tuning on the constructed instruction data and further align the LLMs with human feedback using the DPO algorithm on our cross-lingual human feedback dataset. We evaluate the multilingual understanding and generating capabilities of xLLMs-100 on five multilingual benchmarks. Experimental results show that xLLMs-100 consistently outperforms its peers across the benchmarks by considerable margins, defining a new state-of-the-art multilingual LLM that supports 100 languages.

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。