Guido Ivetta and Luciana Benotti at EMNLP 2025: Presenting HESEIA

The EMNLP 2025 conference was held in Suzhou, China, from November 4th to 9th, 2025.

Guido Ivetta and Luciana Benotti were be in attendance to present their latest research. On Thursday, November 6th, from 10:30 to 12:00 (local time), Guido presented the paper titled: “HESEIA: A Community Dataset for Evaluating Social Biases in Large Language Models, Co-Designed in Real School Environments in Latin America.” He was joined by Luciana, who participated in the Q&A session. Additionally, both Luciana and Guido presented the HESEIA poster in person at the conference.

This work is authored by Guido Ivetta, Marcos J. Gómez, Sofía Martinelli, Pietro Palombini, M. Emilia Echeveste, Nair Carolina Mazzeo, Beatriz Busaniche, and Luciana Benotti.

The HESEIA project emerged from a teacher training course held in 2024, which involved 370 teachers and over 5,000 students from 189 schools across Latin America. This process resulted in the collaborative construction of a dataset comprising 46,499 sentences designed to identify and analyze biases and stereotypes in artificial intelligence models. Unlike other existing resources, HESEIA incorporates an intersectional and situated perspective, reflecting the local contexts and pedagogical experience of the educational communities.