This corpus was compiled by gathering texts addressing the LGBTQIA+ community published on the official website of the Taiwanese government from the 12th of November, 2009 to the 6th of February, 2024. It comprises 211,644 tokens, 175,725 words, and 5,382 sentences. Additionally, it encompasses 26,924 lemmas and 23,325 unique word forms (including non-words). The uploaded file contains both a plain text version (without POS tags or lemmas, but retaining all structures and structural attributes) and a vertical file (presenting the corpus in vertical format, including POS tags, lemmas, structures, and attributes). POS tagging and lemmatization were executed using the Sketch Engine platform (http://www.sketchengine.eu; Kilgarriff et al. 2004, Kilgarriff et al. 2014).