DOSA: A Dataset of Social Artifacts from Different Indian Geographical Subcultures

TL;DR

This paper introduces DOSA, a culturally diverse dataset of social artifacts from Indian subcultures, highlighting the importance of community participation in creating culturally aware benchmarks for language models.

Contribution

It presents the first community-generated dataset of social artifacts from Indian subcultures, using participatory methods to ensure cultural relevance and diversity.

Findings

LLMs show significant variation in understanding artifacts across subcultures

The dataset reveals gaps in LLMs' cultural knowledge

Participatory data collection enhances cultural representation in NLP

Abstract

Generative models are increasingly being used in various applications, such as text generation, commonsense reasoning, and question-answering. To be effective globally, these models must be aware of and account for local socio-cultural contexts, making it necessary to have benchmarks to evaluate the models for their cultural familiarity. Since the training data for LLMs is web-based and the Web is limited in its representation of information, it does not capture knowledge present within communities that are not on the Web. Thus, these models exacerbate the inequities, semantic misalignment, and stereotypes from the Web. There has been a growing call for community-centered participatory research methods in NLP. In this work, we respond to this call by using participatory research methods to introduce $\textit{DOSA}$, the first community-generated $\textbf{D}$ataset $\textbf{o}$f 615 $\textbf{S}$ocial $\textbf{A}$rtifacts, by engaging with 260 participants from 19 different Indian geographic subcultures. We use a gamified framework that relies on collective sensemaking to collect the names and descriptions of these artifacts such that the descriptions semantically align with the shared sensibilities of the individuals from those cultures. Next, we benchmark four popular LLMs and find that they show significant variation across regional sub-cultures in their ability to infer the artifacts.