arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2502.03220cs.CL
收藏

缓解跨语言职位检索中的语言偏见:招聘平台视角

Mitigating Language Bias in Cross-Lingual Job Retrieval: A Recruitment Platform Perspective

Chulalongkorn University(朱拉隆功大学)

Napat Laosaengpha, Thanit Tativannarat, Attapol Rutherford, Ekapol Chuangsuwanich

AI总结:

本文提出统一句子编码器,通过多任务双编码器框架联合学习简历与职位文本的多个组成部分,并引入LBKL指标评估语言偏见,在较小模型规模下实现优于现有模型的跨语言职位检索性能。

摘要:

理解简历和职位发布中的文本组成部分对于提高在线招聘平台中的职位匹配准确性和优化职位搜索系统至关重要。然而,现有工作主要侧重于分析这些信息中的各个独立组成部分,需要多个专用工具来分析每个方面。这种割裂的方法可能会阻碍招聘相关文本处理的整体泛化能力。因此,我们提出了一种统一的句子编码器,利用多任务双编码器框架将多个组成部分联合学习到统一的句子编码器中。结果表明,尽管模型规模较小,我们的方法仍优于其他最先进的模型。此外,我们提出了一种新的度量指标——语言偏见KL散度(LBKL),用于评估编码器中的语言偏见,证明了显著的偏见减少和优越的跨语言性能。

英文摘要:

Understanding the textual components of resumes and job postings is critical for improving job-matching accuracy and optimizing job search systems in online recruitment platforms. However, existing works primarily focus on analyzing individual components within this information, requiring multiple specialized tools to analyze each aspect. Such disjointed methods could potentially hinder overall generalizability in recruitment-related text processing. Therefore, we propose a unified sentence encoder that utilized multi-task dual-encoder framework for jointly learning multiple component into the unified sentence encoder. The results show that our method outperforms other state-of-the-art models, despite its smaller model size. Moreover, we propose a novel metric, Language Bias Kullback-Leibler Divergence (LBKL), to evaluate language bias in the encoder, demonstrating significant bias reduction and superior cross-lingual performance.

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。

↑