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 2407.03536cs.CL
收藏

孟加拉语大语言模型中的社会偏见:性别与宗教偏见的实证研究

Social Bias in Large Language Models For Bangla: An Empirical Study on Gender and Religious Bias

Bangladesh University of Engineering and Technology (BUET)(孟加拉工程技术大学)

Jayanta Sadhu, Maneesha Rani Saha, Rifat Shahriyar

AI总结:

针对孟加拉语大语言模型偏见研究匮乏的问题,该研究考察其性别与宗教两类社会偏见,构建偏见测量基准数据集,测试两种探测技术,为该领域首项同类研究,资源公开可用。

摘要:

大语言模型(LLMs)的快速发展使得偏见研究成为关键领域。评估LLMs中嵌入的各类偏见的影响,对于确保其在敏感领域的公平使用至关重要。尽管针对英语的偏见评估已有大量研究,但对于孟加拉语这类主要语言,相关研究仍十分匮乏。本研究考察了孟加拉语LLM生成输出中的两类社会偏见。本文的主要贡献包括:(1)针对孟加拉语的两类不同社会偏见开展偏见研究;(2)构建了一个用于偏见测量基准测试的精选数据集;(3)测试了两种适用于孟加拉语语境下偏见检测的探测技术。据我们所知,这是首项针对孟加拉语LLM偏见评估的此类研究。我们的所有代码与资源均公开可用,以推动孟加拉语自然语言处理(NLP)领域偏见相关研究的发展。

英文摘要:

The rapid growth of Large Language Models (LLMs) has put forward the study of biases as a crucial field. It is important to assess the influence of different types of biases embedded in LLMs to ensure fair use in sensitive fields. Although there have been extensive works on bias assessment in English, such efforts are rare and scarce for a major language like Bangla. In this work, we examine two types of social biases in LLM generated outputs for Bangla language. Our main contributions in this work are: (1) bias studies on two different social biases for Bangla, (2) a curated dataset for bias measurement benchmarking and (3) testing two different probing techniques for bias detection in the context of Bangla. This is the first work of such kind involving bias assessment of LLMs for Bangla to the best of our knowledge. All our code and resources are publicly available for the progress of bias related research in Bangla NLP.

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。

↑