Multilingual CheckList: Generation and Evaluation
This work addresses the problem of limited multilingual evaluation capabilities for NLP researchers, offering an incremental improvement by automating template generation to reduce reliance on native speakers.
The paper tackled the challenge of scaling CheckList-based evaluation to multiple languages by developing an automated Template Extraction Algorithm (TEA) to generate templates from machine translations, comparing it with human-created versions across 10 languages including Hindi, and found that TEA with human verification is ideal for scaling while providing good performance estimates.
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.