AmericasnLP 2025: Machine Translation Metrics for Indigenous Languages

AmericasNLP 2025 invites researchers to develop machine translation metrics for Indigenous languages, focusing on Guarani, Bribri, and Nahuatl.

The AmericasNLP 2025 Shared Task invites researchers to participate in a competition focused on advancing automatic evaluation metrics for machine translation, specifically targeting Indigenous languages. This task aims to address the unique challenges posed by these languages, such as polysynthesis and lack of orthographic standardization, which are not well-handled by traditional metrics like BLEU and ChrF.

Participants will develop metrics to assess the quality of translations from Spanish into Guarani, Bribri, and Nahuatl. The goal is to foster innovation in NLP and improve the evaluation of machine translation systems for underrepresented languages.

For more details, visit the shared task description page. To register, fill out the registration form.

The AmericasNLP workshop will be held in Albuquerque, New Mexico, on May 4, 2025, as part of NAACL 2025.

Tags: AmericasNLP, Machine Translation, Indigenous Languages, NLP, Metrics, Guarani, Bribri, Nahuatl