Preserving Grammatical Gender when Debiasing Word Embeddings in Spanish
Accepted for publication in Issue No. 75 of the SEPLN journal and for presentation at the XLI International SEPLN Conference, 2025.
In this paper, we propose INLP-Gram, an algorithm designed to mitigate gender bias in Spanish word embeddings while preserving grammatical gender information. It is an adaptation of the Iterative Nullspace Projection (INLP).