Loading...
Loading...
Available on 1 platform
Sign in to view source links and access this dataset
A benchmark dataset for evaluating pragmatic alignment in language models, focusing on the Rioplatense Spanish dialect. The dataset was created by somosnlp-hackathon-2026 and was last updated on June 11, 2026. It is designed to address the tendency of models to interpret language too literally and over-moderate colloquial expressions from specific linguistic communities.
License is unknown; terms of use must be verified before application.