While generative multilingual models are rapidly being deployed, their safety and fairness
evaluations are largely limited to resources collected in English. This is especially …

Studying bias detection and mitigation methods in natural language processing and the
particular case of machine translation is highly relevant, as societal stereotypes might be …

This paper introduces the GenderQueer Test Suite, an evaluation set for assessing machine
translation (MT) systems' capabilities in handling gender-diverse and queer-inclusive …

This article examines the lexicographic material found in Dictionary of Contemporary
Icelandic (OMI). We focus on four processes for acquiring new material: 1) corpus data, 2) …

We assess the difficulty of gender resolution in literary-style dialogue settings and the
influence of gender stereotypes. Instances of the test suite contain spoken dialogue …

All language data are inherently biased, as collection methods, availability of texts and
recordings, and the views of the collectors will always affect the process and its results. We …

Large language models, trained on vast datasets, exhibit increased output quality in
proportion to the amount of data that is used to train them. This data-driven learning process …