PorTEXTO: A European Portuguese Benchmark for Visual Text Extraction
João Cardeira, Diogo Glória-Silva, Manuel Letras da Luz, Rafael Ferreira, Diogo Tavares, David Semedo, João Magalhães
Abstract
European Portuguese (pt-PT) is largely absent from OCR benchmarks, which skew toward high-resource languages. The few benchmarks that cover pt-PT focus on historical artifacts and literature. This work addresses modern OCR applications, introducing PorTEXTO, the first benchmark for contemporary and culturally relevant pt-PT visual text extraction. To ascertain quality, we employ an annotation pipeline combining transcriptions from a frontier LVLM with exhaustive review by native speakers. We observe a sharp performance drop from synthetic to real world samples in most models, and find that, currently, specialized multilingual data is a better driver for pt-PT performance than model size or resolution budget, motivating the release of open pt-PT OCR resources.
Topics
Classified with taxonomy v2 on Wed, 2 Sept 2026.