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 2203.12865cs.CLcs.LG
收藏

多语种CheckList的生成与评价

Multilingual CheckList: Generation and Evaluation

Duke University(杜克大学);Microsoft Research(微软研究院);IIT Kharagpur(印度理工学院克勒格布尔分校);Microsoft R&D(微软研发中心)

Karthikeyan K, Shaily Bhatt, Pankaj Singh, Somak Aditya, Sandipan Dandapat, Sunayana Sitaram, Monojit Choudhury

摘要:

多语言评估基准通常包含有限的高资源语言,并且不测试特定语言能力的模型。CheckList是一种基于模板的评估方法,用于测试模型的特定功能。CheckList模板创建过程需要母语人士,这对扩展到数百种语言构成了挑战。在这项工作中,我们探索多种方法来生成多语言检查表。本文提出了一种从机器翻译的源语言模板实例中自动提取目标语言CheckList模板的算法--模板提取算法(TEA)。我们比较了TEA检查表与通过不同级别的人为干预创建的检查表。我们进一步引入度量沿维度的成本,多样性,效用,和正确性比较检查表。我们彻底分析了在印地语中创建检查列表的不同方法。此外,我们还尝试了9种不同的语言。我们发现,TEA其次是人工验证是理想的扩展基于检查表的评估多种语言,而TEA提供了一个很好的估计模型的性能。

英文摘要:

Multilingual evaluation benchmarks usually contain limited high-resource languages and do not test models for specific linguistic capabilities. CheckList is a template-based evaluation approach that tests models for specific capabilities. The CheckList template creation process requires native speakers, posing a challenge in scaling to hundreds of languages. In this work, we explore multiple approaches to generate Multilingual CheckLists. We device an algorithm - Template Extraction Algorithm (TEA) for automatically extracting target language CheckList templates from machine translated instances of a source language templates. We compare the TEA CheckLists with CheckLists created with different levels of human intervention. We further introduce metrics along the dimensions of cost, diversity, utility, and correctness to compare the CheckLists. We thoroughly analyze different approaches to creating CheckLists in Hindi. Furthermore, we experiment with 9 more different languages. We find that TEA followed by human verification is ideal for scaling Checklist-based evaluation to multiple languages while TEA gives a good estimates of model 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。

↑