Overview of ADoBo at IberLEF 2025: Automatic Detection of Anglicisms in Spanish

TL;DR

This paper reports on ADoBo 2025, a shared task focused on detecting English borrowings in Spanish texts, comparing various models including LLMs and rule-based systems with performance variability.

Contribution

It provides an overview of the shared task, the participating systems, and their performance, highlighting the challenges in anglicism detection in Spanish.

Findings

Performance ranged from F1 0.17 to 0.99 across systems.

Deep learning and Transformer models showed high variability in results.

The task highlights the complexity of automatic anglicism detection.

Abstract

This paper summarizes the main findings of ADoBo 2025, the shared task on anglicism identification in Spanish proposed in the context of IberLEF 2025. Participants of ADoBo 2025 were asked to detect English lexical borrowings (or anglicisms) from a collection of Spanish journalistic texts. Five teams submitted their solutions for the test phase. Proposed systems included LLMs, deep learning models, Transformer-based models and rule-based systems. The results range from F1 scores of 0.17 to 0.99, which showcases the variability in performance different systems can have for this task.