Overview of HISEMOTIONS at IberLEF 2026: Historical Text-Based Emotion Detection in Early Modern Spanish Correspondence

Resumen

This paper presents the results of HISEMOTIONS, a shared task at IberLEF 2026 focused on Text-Based Emotion Detection in Early Modern Spanish epistolary texts from the 16th and 17th centuries. The task aims to foster the development and evaluation of computational methods specifically adapted to historical Spanish language varieties, addressing the semantic and diachronic challenges of expressing emotion and affective states in Early Modern Spanish correspondence. The core challenge proposed within HISEMOTIONS is Multi-label Emotion Detection, oriented toward predicting the emotion perceived by a reader when interpreting an author's text. The final ranking encompasses the results of 7 teams, each of which proposed a distinct approach to the problem.