Bridging the Digital Divide: Performance Variation across Socio-Economic Factors in Vision-Language Models
This addresses the problem of AI performance disparities for low-income groups, which are often overlooked in model evaluations, making it an incremental contribution by focusing on a specific under-represented domain.
The study evaluated CLIP's performance on a geo-diverse dataset (Dollar Street) and found that performance is consistently lower for poorer households compared to wealthier ones across various topics and countries, highlighting performance inequality based on income levels.
Despite the impressive performance of current AI models reported across various tasks, performance reports often do not include evaluations of how these models perform on the specific groups that will be impacted by these technologies. Among the minority groups under-represented in AI, data from low-income households are often overlooked in data collection and model evaluation. We evaluate the performance of a state-of-the-art vision-language model (CLIP) on a geo-diverse dataset containing household images associated with different income values (Dollar Street) and show that performance inequality exists among households of different income levels. Our results indicate that performance for the poorer groups is consistently lower than the wealthier groups across various topics and countries. We highlight insights that can help mitigate these issues and propose actionable steps for economic-level inclusive AI development. Code is available at https://github.com/MichiganNLP/Bridging_the_Digital_Divide.