Few-Shot Cross-Lingual Transfer for Prompting Large Language Models in Low-Resource Languages

📄 arXiv: 2403.06018v1 📥 PDF

作者: Christopher Toukmaji

分类: cs.CL, cs.AI, cs.LG

发布日期: 2024-03-09

备注: 47 pages, 26 figures; a thesis submitted in partial satisfaction of the requirements for the degree of Bachelor of Science in Computer Science at the University of California - Santa Cruz


💡 一句话要点

提出跨语言少样本学习方法以提升低资源语言的提示能力

🎯 匹配领域: 支柱九:具身大模型 (Embodied Foundation Models)

关键词: 预训练语言模型 少样本学习 跨语言迁移 低资源语言 自然语言处理 提示学习 语言自适应微调

📋 核心要点

  1. 现有的预训练语言模型主要基于英语语料,导致低资源语言的适应性不足,影响了其在多语言环境中的应用。
  2. 论文提出通过少样本提示、语言自适应微调和神经机器翻译三种方法,来提升PLMs在低资源语言中的提示能力。
  3. 实验结果显示,少样本提示方法在多项任务上表现优于其他方法,且在计算效率上具有显著优势。

📝 摘要(中文)

大型预训练语言模型(PLMs)在自然语言处理领域取得了显著进展,尤其是在“提示”或上下文学习方面。然而,现有的PLMs通常以英语语料为主,导致其他语言的适应性不足。本文评估了如何将以英语为主的LLaMa模型适应于低资源语言(如基尼亚卢旺达语、豪萨语和卢干达语)的提示任务。我们比较了三种方法:少样本提示、语言自适应微调和神经机器翻译,结果表明,少样本提示在所有任务和语言上表现最佳,且在计算效率上优于其他方法。

🔬 方法详解

问题定义:本文旨在解决如何将大型预训练语言模型有效适应于低资源语言的提示任务。现有方法多依赖于英语训练,导致其他语言的模型性能不足。

核心思路:论文提出通过少样本提示、语言自适应微调(LAFT)和神经机器翻译三种方法,探索在低资源语言中进行有效的提示学习。少样本提示被认为是计算效率最高的方法。

技术框架:整体架构包括三个主要模块:少样本提示模块、语言自适应微调模块和神经机器翻译模块。每个模块针对不同的任务进行优化,最终通过实验评估其在低资源语言中的表现。

关键创新:最重要的创新在于发现少样本提示在多项任务上优于语言自适应微调和翻译方法,挑战了传统上认为微调是最佳选择的观点。

关键设计:在实验中,采用了不同的提示设计和参数设置,确保模型在低资源语言上的适应性和性能提升。具体的损失函数和网络结构细节在论文中进行了详细描述。

🖼️ 关键图片

fig_0
fig_1
fig_2

📊 实验亮点

实验结果表明,少样本提示方法在所有任务和语言上表现优于语言自适应微调和神经机器翻译,且在计算效率上具有显著优势。具体而言,少样本提示在各项任务中均显示出统计显著性提升,验证了其在低资源语言中的有效性。

🎯 应用场景

该研究的潜在应用领域包括多语言自然语言处理、跨语言信息检索和低资源语言的机器翻译等。通过提升低资源语言的模型性能,能够更好地服务于全球多样化的语言需求,促进语言平等和信息获取的公平性。

📄 摘要(原文)

Large pre-trained language models (PLMs) are at the forefront of advances in Natural Language Processing. One widespread use case of PLMs is "prompting" - or in-context learning - where a user provides a description of a task and some completed examples of the task to a PLM as context before prompting the PLM to perform the task on a new example. Only the largest, most capable PLMs are able to perform in-context learning effectively, and these models are typically trained with a predominantly English corpus, leaving all other languages behind. The data limitations in most languages preclude the training of language-specific PLMs capable of prompting. Albeit the surge in work of prompting settings, it is still unclear how PLMs should be adapted cross-lingually specifically for prompting. We evaluate the possible methods to adapt LLaMa, a 7B parameter open-source PLM mainly trained in English, for prompting in low-resource languages, namely for Kinyarwanda, Hausa, and Luganda. We consider three methods: few-shot prompting (prompt), language-adaptive fine-tuning (LAFT), and neural machine translation (translate), and evaluate on abstractive summarization, multi-class topic classification, and named-entity recognition. Although LAFT carries the greatest compute cost and intuitively should lead to the best results, our experiments exhibit that LAFT is only occasionally the optimal choice for adapting PLMs for prompting. Rather, the translate and prompt settings are a compute-efficient and cost-effective method of few-shot prompting for the selected low-resource languages. We find that the results are task and language dependent but find that the prompting method is the best on average across all tasks and languages. Results show that the prompt setting performs better than both translating and LAFT with statistical significance for all shots when aggregated across all tasks and languages.