EtiCor: Corpus for Analyzing LLMs for Etiquettes
This addresses the need for better evaluation of LLMs on culturally diverse social norms, though it is incremental as it introduces a new dataset and task.
The authors tackled the problem of evaluating large language models (LLMs) on region-specific etiquettes by proposing EtiCor, a corpus of social norms from five global regions, and found that LLMs mostly fail to understand etiquettes from non-Western regions.
Etiquettes are an essential ingredient of day-to-day interactions among people. Moreover, etiquettes are region-specific, and etiquettes in one region might contradict those in other regions. In this paper, we propose EtiCor, an Etiquettes Corpus, having texts about social norms from five different regions across the globe. The corpus provides a test bed for evaluating LLMs for knowledge and understanding of region-specific etiquettes. Additionally, we propose the task of Etiquette Sensitivity. We experiment with state-of-the-art LLMs (Delphi, Falcon40B, and GPT-3.5). Initial results indicate that LLMs, mostly fail to understand etiquettes from regions from non-Western world.