Santiago Góngora

Master's student

Hi! I'm Santiago Góngora, a Master's student at Grupo de Procesamiento de Lenguaje natural (🇺🇾), advised by Luis Chiruzzo (Universidad de la República, Uruguay) and Gonzalo Méndez (Universidad Complutense de Madrid, Spain). I'm currently working on the modelling of the linguistical and creative complexities of an RPG's game master.

Please don't hesitate to contact me for any questions, comments or requests.

Updates

Research Interests and Selected Publications

From a very early age I was deeply interested in arts, and how humans used them to communicate their legacy in different forms. Born in the 90s, it became natural to me to think of games and computers as a platform for building interactive experiences that let humans communicate ideas and feelings. That’s why I'm passionate about Natural Language Processing (NLP) and its applications. These are my main research interests, but probably I am interested in many more.

Interactive Storytelling and Role-playing Games

As a Master's student I am working in Interactive Storytelling. I'm primarily interested in modelling the many abilities that a human game master has to use when running a RPG campaign.

  • Góngora, S., Chiruzzo, L., Méndez, G., & Gervás, P. (2024, June). PAYADOR: A Minimalist Approach to Grounding Language Models on Structured Data for Interactive Storytelling and Role-playing Games [Paper] [Repository]
  • Góngora, S., Chiruzzo, L., Méndez, G., & Gervás, P. (2023, November). Skill check: Some considerations on the evaluation of gamemastering models for role-playing games [Paper] [Repository]

Low-resource languages - Guarani-Spanish Machine Translation

For my Bachelor thesis we built some resources for Guarani, an indigenous language from South america, considered a low-resource language in the NLP community. Since then, I worked on different projects related to Guarani, such as lexical semantics, code-switching detection and machine translation. Recently we presented at NAACL2024 an approach to tackle the lack of data in Guarani-Spanish MT. The paper was awarded with the Theme track paper award!

  • Lucas, A., Baladón, A., Pardiñas, V., Agüero-Torales, M., Góngora, S., & Chiruzzo, L. (2024, June). Grammar-based Data Augmentation for Low-Resource Languages: The Case of Guarani-Spanish Neural Machine Translation [Paper] [Repository]
  • Chiruzzo, L., Agüero-Torales, M., Giménez-Lugo, G., Alvarez, A., Rodríguez, Y., Góngora, S., & Solorio, T. (2023, September). Overview of GUA-SPA at IberLEF 2023: Guarani-Spanish Code Switching Analysis [Paper] [Repository]
  • Chiruzzo, L., Góngora, S., Alvarez, A., Giménez-Lugo, G., Agüero-Torales, M., & Rodríguez, Y. (2022, June). Jojajovai: A parallel guarani-spanish corpus for mt benchmarking [Paper] [Repository]
  • Góngora, S., Giossa, N., & Chiruzzo, L. (2021, June). Experiments on a Guarani corpus of news and social media [Paper] [Repository]

Games + NLP in education

As part of a collaboration between our research group and Políticas lingüísticas-ANEP (the government agency responsible for the Uruguayan public educational system), we are working on games to enhance English teaching in primary schools, mainly for those located in rural areas.

  • Chiruzzo, L., Musto, L., Góngora, S., Carpenter, B., Filevich, J., & Rosá, A. (2022, December). Using nlp to support english teaching in rural schools [Paper]