Unsupervised Automatic Short Answer Grading and Essay Scoring: A Weakly Supervised Explainable Approach
A method for unsupervised automatic short answer grading and unsupervised automatic essay scoring, that is competitive with LLM, with way less parameters, that is white box with interpretable features.
Jun 5, 2025
(Preprint) StandUp4AI: A New Multilingual Dataset for Humor Detection in Stand-up Comedy Videos
StandUp4AI is a new multimodal dataset for humor detection, featuring over 330 hours of stand-up comedies in seven languages. This dataset surpasses existing ones in size and diversity, and experiments show it's a valuable resource for training models, with potential for improved performance when combined with enhanced audio speech recognition methods.
Jun 5, 2025
StandUpComedy🎤😄👏
Humor Detection and Modelization in Multimodal Natural Language for Embodied Conversational Agents.
May 15, 2025
[Tesis pre/postgrado] [Pagada] Change my view! -- Analisis de argumentacion multimodal
Tesista pagado sobre un tema social interesante en un proyecto de investigacion!
Jan 1, 2025
[FOUND] [Tesis pre/postgrado] [Pagada] Xenophobias -- Deteccion y reduccion de sesgos etnicos en LLM
Tesista pagado sobre un tema social interesante en un proyecto de investigacion!
Jan 1, 2025
Fondecyt de Iniciacion🗣️💬🤖
Multimodal Argumentation Mining in Groups Assisted by an Embodied Conversational Agent
Dec 31, 2024
Large Multimodal Models @ CENIAMODAL
We organized the CENIAMODAL workshop, where I gave a talk on Large Multimodal Models.
Dec 17, 2024
Tackling Biases In or Using Generative AI @ JSIC'24
Tackling Biases In or Using Generative AI
Dec 4, 2024
A Study of Nationality Bias in Names and Perplexity using Off-the-Shelf Affect-related Tweet Classifiers
We've developed a method to measure biases in AI models related to named entities from different countries, and our results show that the presence of certain country names can significantly influence predictions, such as hate speech detection and emotion analysis, with changes of up to 23% and 60% respectively! Our findings suggest that these biases are rooted in the pre-training data of language models, and we've uncovered interesting patterns that reveal how the language and country of origin can impact model predictions, with English-speaking country names having a particularly strong effect.
Nov 1, 2024
[FOUND] [Tesis pre/postgrado] JAJAJJJJJ -- Deteccion de humor en videos de stand-up comedy
Tesista sobre un tema social interesante en un proyecto de investigacion!
Nov 1, 2024