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 2611.06706cs.CLcs.AIcs.CYcs.LG
收藏

ALTO:非洲与黎凡特地区技术促成的基于性别的暴力语料库

ALTO: African and Levantine Tech-Facilitated Gender-Based Violence Corpus

Meedan;Oxford Internet Institute, University of Oxford(牛津大学互联网研究所)

Tianwei Zhang, Ivy Njoki Nyagwencha, Zahraa Dirani, Shalini Joshi, Scott A. Hale

AI总结:

该研究构建了ALTO语料库,与CSO合作众包收集三种非全球北方语言的TFGBV实例,用主动学习标注后经实验发现现有模型在该语料库上TFGBV召回率低,微调可改善,还发布了相关资源。

摘要:

技术促成的基于性别的暴力(TFGBV)是一种普遍存在的网络伤害,通过多种直接和隐蔽形式影响女性及性别少数群体。当前用于性别伤害检测的模型和标注数据集,主要由来自全球北方语言的数据以及在这些语言上训练的模型主导。我们与来自约旦、西非法语区和肯尼亚的民间社会组织(CSO)合作,采用社区主导的方法,众包收集黎凡特阿拉伯语、非洲法语和斯瓦希里语中具有文化和语言相关性的TFGBV实例。我们使用主动学习对众包数据进行标注,构建了一个规模较小但性能良好的数据集(N=3022),命名为ALTO,以改进这三种语言的TFGBV检测。通过基准测试和微调实验,我们发现现有有害内容检测模型在ALTO上的TFGBV召回率较低,经微调后有所提升;模型常遗漏所有语言中的隐蔽伤害,且斯瓦希里语的错误率明显更高。我们报告了实验结果,并发布了标注分类法、数据集和训练后的模型。

英文摘要:

Tech-facilitated gender based violence (TFGBV) is a pervasive online harm affecting women and gender minority groups through its many direct and covert forms. Current models and annotated datasets for gendered harm detection are dominated by data from, and models trained in, Global North languages. In collaboration with civil society organizations (CSOs) from Jordan, Francophone West Africa, and Kenya, we developed a community-led approach to crowdsource culturally and linguistically relevant instances of TFGBV in Levantine Arabic, African French, and Swahili. We annotated the crowdsourced data using active learning to develop a small but performant dataset (N=3022) called ALTO to improve TFGBV detection in the three languages. Through benchmarking and fine-tuning experiments, we show that existing harmful content detection models show poor TFGBV recall within ALTO and improve with fine-tuning. We found that models often miss covert harms across all languages, while error rates are markedly higher in Swahili. We report the experimental results and release the annotation taxonomy, dataset, and trained models.

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。

↑