Documenting Geographically and Contextually Diverse Data Sources: The BigScience Catalogue of Language Data and Resources
This addresses data rights and transparency issues for researchers and practitioners in NLP, though it is incremental as it focuses on documentation rather than new modeling techniques.
The paper tackles the problem of insufficient documentation in large-scale data collection for language models by presenting a methodology for a documentation-first, human-centered approach, resulting in an online catalogue with metadata for geographically diverse language groups.
In recent years, large-scale data collection efforts have prioritized the amount of data collected in order to improve the modeling capabilities of large language models. This prioritization, however, has resulted in concerns with respect to the rights of data subjects represented in data collections, particularly when considering the difficulty in interrogating these collections due to insufficient documentation and tools for analysis. Mindful of these pitfalls, we present our methodology for a documentation-first, human-centered data collection project as part of the BigScience initiative. We identified a geographically diverse set of target language groups (Arabic, Basque, Chinese, Catalan, English, French, Indic languages, Indonesian, Niger-Congo languages, Portuguese, Spanish, and Vietnamese, as well as programming languages) for which to collect metadata on potential data sources. To structure this effort, we developed our online catalogue as a supporting tool for gathering metadata through organized public hackathons. We present our development process; analyses of the resulting resource metadata, including distributions over languages, regions, and resource types; and our lessons learned in this endeavor.