Hora de la Ciudad de México (UTC-6)
15h00 Registro
15h20 Bienvenida
15h30 Overview HOMO-MEX: Hate speech detection in Online Messages directed tOwards the MEXican spanish speaking LGBTQ+ population
Helena Gómez-Adorno / IIMAS
15h45 Intervenciones de participantes en esta tarea
LabTL-INAOE at HOMO-MEX 2024: Distance-based Representations for LGBT+ Phobia Detection
Metztli Ramírez / INAOE
HomoCIC at HOMO-MEX 2024: Deep Learning Approaches for Classifying Homophobic Content in Tweets and Songs: Leveraging LLM and NL
Omar Garcia Vazquez, Marco Cardoso-Moreno, José Alberto Torres-León, Diana Jiménez / CIC
16h30 Overview of DIMEMEX: Detection of Inappropriate Memes from Mexico
Horacio Jesús Jarquín Vásquez / INAOE
16h45 Intervenciones de participantes en esta tarea
TC at DIMEMEX: When hate goes Viral: Detection of Hate Speech in Mexican Memes Using Transformers
Víctor Manuel Bátiz / IT Culiacán
17h00 pausa
17h15 Overview of HOPE - Approaching Hope Speech Detection in Social Media from Two Perspectives, for Equality, Diversity and Inclussion and as Expectations
Fazlourrahman Balouchzahi / CIC
17h30 Intervenciones de participantes en esta tarea
HOPE: A Multilingual Approach to Identifying Positive Communication in Social Media
Fida Ullah, CIC
HOPE2024@IberLEF: A Cross-Linguistic Exploration of Hope Speech Detection in Social Media
Mikhail Krasitskii, CIC
Hope Speech in Social Media Texts using Transformer
Girma Yohannis, Tecnológico de Monterrey
18h15 Otras experiencias
The iimasNLP team at IberAuTexTification 2024: Integrating Graph Neural Networks, Multilingual LLMs, and Stylometry for Automatic Text Identification
Ricardo Loth Zavala-Reyes, IIMAS
Team qIIMAS on Task 2 – Clustering: Notebook for the QuantumCLEF Lab at CLEF 2024
Jorge Téllez-Torres y Javier Tovar-Cortes, IIMAS
GIL-IIMAS UNAM at SemEval-2024 Task 1: SAND: An In Depth Analysis of Semantic Relatedness Using Regression and Similarity Characteristics.
Francisco Lopez-Ponce, Ángel Cadena, Karla Salas-Jiménez, Gemma Bel-Enguix, and David Preciado-Márquez.
PCICUNAM at WASSA 2024: Cross-lingual Emotion Detection Task with Hierarchical Classification and Weighted Loss Functions.
Helena Gómez-Adorno, Jesus Vázquez-Osorio, Gerardo Sierra, Gemma Bel-Enguix.
JK_PCIC_UNAM at CheckThat! 2024: Analysis of Subjectivity in News Sentences Using Transformers-Based Models.
Karla Salas-Jimenez, Iván Díaz, Helena Gómez-Adorno,Gemma Bel-Enguix and Gerardo Sierra
19h30 Cierre
Hora de la Ciudad de México (UTC-6)
08h00 Registro
08h50 Bienvenida
09h00 1ra sesión pláticas generales (pláticas de 25 minutos incluyendo 5 minutos para preguntas)
Text2graph: An API and Web App to Transform Texts into Different Graph Representations
Helena Gómez Adorno / IIMAS
Computational Study of Dream Interpretations: Psychoanalytic Human vs Artificial Analyses
Hiram Calvo / CIC
Analysis of systems’ performance in natural language processing competitions
Sergio Nava / INFOTEC-INAOE
Alucinaciones en LLM: Descripción e invitación a la Experiencia Mu-SHROOM (SemEval 2025).
Fernando Sánchez / CIMAT
11h00 pausa
11h15 Conferencia Magistral
Best Frenemies Forever: Can Therapists and AI Models Trust Each Other?
Esaú Villatoro Tello / IDIAP, Suiza
Abstract:
In this talk, I’ll describe a case study of automatic depression detection from clinical interviews. As known, automatic depression detection from conversational data has gained significant interest in recent years. The DAIC-WOZ dataset, interviews conducted by a human-controlled virtual agent, has been widely used for this task. However, very recent studies report surprisingly high performance when incorporating the interviewer's prompts into the model training. Through various experiments, I will show that relying on an AI model that uses all the information provided by a therapist may not be the best approach; therefore, therapists should be cautious in fully trusting AI models trained this way. How can we make these trust issues disappear? I will discuss and share insights and ideas toward this goal aimed at fostering a more reliable integration of AI in clinical settings.
Short bio:
Dr. Villatoro-Tello has been working as a Researcher in the Speech and Audio Processing Group at the Idiap Research Institute in Switzerland since September 2019. From June 2012 to December 2022, he held a tenured position at the Universidad Autónoma Metropolitana campus Cuajimalpa (UAM-C) in Mexico City. Dr. Villatoro-Tello has over 13 years of experience in the Computer Science field. He has been part of the main organisation committee of the Mexican NLP Summer School, co-located with NAACL during the years 2021 and 2024. Dr. Villatoro-Tello is a regular member of the Program Committee of many top NLP conferences, e.g., ACL, EACL, EMNLP, COLING, NAACL, LREC, CoNLL, IJCNLP, and CLEF, and serves regularly as Area Chair of the ACL Rolling Review System. He has supervised the thesis work of more than 28 students (among Bachelors, Masters, and PhD). He has authored more than a hundred publications among conference papers, journal papers and book editions. Dr Villatoro-Tello has been the coordinator and principal investigator (PI) of more than five projects funded by the Mexican National Council of Research (CONACyT), and very recently, he was awarded by the Swiss National Foundation Science (SNSF) with a grant to work on novel AI technologies applied to mental health and well-being domains (2025-2029). Overall, his primary research interests include the areas of Computational Linguistics (CL), Information Retrieval (IR), and Natural Language Processing (NLP). His most recent work lies in the intersection of NLP and Social Sciences, with special emphasis on applied NLP in the Mental Health domain, Authorship Analysis (AA) and Natural and Spoken Language Understanding (NLU/SLU), Semantic Analysis (SA) and News Media Profiling.
12h15 pausa
12h30 2da sesión pláticas generales (pláticas de 25 minutos incluyendo 5 minutos para preguntas)
El estado de la traducción automática para lenguas originarias de México
Iván Meza / IIMAS
PLN Multilingüe: abrazando la diversidad
Ximena Gutierrez / UNAM
Natural Language Understanding for Navigation of Service Robots in Low-Resource Domains and Languages: Scenarios in Spanish and Nahuatl
Rosa Ortega / UAEH
Texto a voz para narraciones infantiles
Carmen V. Robles Contreras / BUAP
14h30 Comida