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 2305.05940cs.CL
收藏

多语言LLM是对齐下更好的跨语言上下文学习器

Multilingual LLMs are Better Cross-lingual In-context Learners with Alignment

DTU(德里技术大学);IIT Delhi(印度德里理工学院)

Eshaan Tanwar, Subhabrata Dutta, Manish Borthakur, Tanmoy Chakraborty

AI总结:

针对跨语言文本分类中随机提示构建因输入输出空间缺乏对齐而受限的问题,提出X-InSTA策略,通过注入语义连贯性和跨语言任务对齐,在3个任务44个语言对上大幅超越随机选择。

摘要:

上下文学习(ICL)随着大型语言模型能够在无任何梯度更新的情况下,基于少量标注样本推断测试标签而展开。支持ICL的大型语言模型为在低资源环境下绕过反复的标注成本提供了有希望的一步。然而,过去仅有少数研究探讨了跨语言环境下的ICL,在这种环境中,将标签知识从高资源语言转移到低资源语言的需求极为关键。为弥补这一差距,我们首次对跨语言文本分类的ICL进行了深入分析。我们发现,在跨语言ICL的情况下,选择随机输入-标签对来构建提示上下文的主流模式存在严重局限,主要原因是输入空间和输出空间缺乏对齐。为缓解此问题,我们提出了一种新颖的提示构建策略——跨语言上下文源-目标对齐(X-InSTA)。通过在输入示例的语义中注入连贯性,并在源语言和目标语言之间实现基于任务的对齐,X-InSTA在使用44个不同跨语言对的三个不同任务中,能够以巨大优势超越随机提示选择。

英文摘要:

In-context learning (ICL) unfolds as large language models become capable of inferring test labels conditioned on a few labeled samples without any gradient update. ICL-enabled large language models provide a promising step forward toward bypassing recurrent annotation costs in a low-resource setting. Yet, only a handful of past studies have explored ICL in a cross-lingual setting, in which the need for transferring label-knowledge from a high-resource language to a low-resource one is immensely crucial. To bridge the gap, we provide the first in-depth analysis of ICL for cross-lingual text classification. We find that the prevalent mode of selecting random input-label pairs to construct the prompt-context is severely limited in the case of cross-lingual ICL, primarily due to the lack of alignment in the input as well as the output spaces. To mitigate this, we propose a novel prompt construction strategy -- Cross-lingual In-context Source-Target Alignment (X-InSTA). With an injected coherence in the semantics of the input examples and a task-based alignment across the source and target languages, X-InSTA is able to outperform random prompt selection by a large margin across three different tasks using 44 different cross-lingual pairs.

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。

↑