Mario R-C

My research interests are NLP and Dialogue Management (e.g., sentiment analysis and generative conversational agents).

Cutting Through Overload: Efficient Token Dropping for Speech Emotion Recognition in Multimodal Large Language Models
J. Bellver-Soler, Mario Rodríguez-Cantelar, R. Córdoba, L.F. D’Haro
IWSDS 2025, May 2025
paper

Cvqa: Culturally-diverse multilingual visual question answering benchmark
D. Romero, C. Lyu, H.A. Wibowo, T. Lynn, I. Hamed, A.N. Kishore, ... Mario Rodríguez-Cantelar, A.F. Aji
arXiv preprint arXiv:2406.05967
paper
Multilingual Speech Emotion Recognition combining Audio and Text using Small Language Models
J. Bellver, Mario Rodríguez-Cantelar, M. Estecha Garitagoitia, R. De Córdoba Herralde, L.F. D’Haro
IberSPEECH 2024, pp. 211–215
paper

Overview of Robust and Multilingual Automatic Evaluation Metrics for Open-Domain Dialogue Systems at DSTC11 Track 4
Mario Rodríguez-Cantelar, C. Zhang, C. Tang, K. Shi, S. Ghazarian, J. Sedoc, ... A. Rudnicky
Proceedings of the 11th Dialog System Technology Challenge (DSTC11), pp. 260–273
paper
Automatic Detection of Inconsistencies and Hierarchical Topic Classification for Open-Domain Chatbots
Mario Rodríguez-Cantelar, M. Estecha-Garitagoitia, L.F. D’Haro, F. Matía, R. Córdoba
Applied Sciences, 13(16), 9055
paper
THAURUS: An Innovative Multimodal Chatbot Based on the Next Generation of Conversational AI
M. Estecha-Garitagoitia, Mario Rodríguez-Cantelar, A.G. Ruiz, C.G.F. García, S.E. Romero, C. Conforto, ... L.F. D’Haro
Alexa Prize SocialBot Grand Challenge 5
paper
An Emotional Model Based on Fuzzy Logic and Social Psychology for a Personal Assistant Robot
G. Fernández-Blanco Martín, F. Matía, L. García Gómez-Escalonilla, D. Galán, M.G. Sánchez-Escribano, P. de la Puente, Mario Rodríguez-Cantelar
Applied Sciences, 13(5), 3284
paper

Automatic Detection of Inconsistencies in Open-Domain Chatbots
J.M. Prats, M. Estecha-Garitagoitia, Mario Rodríguez-Cantelar, L. Fernando
Proceedings of IberSPEECH 2022, pp. 116–120
paper

Genuine2: An Open Domain Chatbot Based on Generative Models
Mario Rodríguez-Cantelar, D. de la Cal, M. Estecha, A.G. Gutiérrez, D. Martin, N.R.N. Milara, ... L.F. D’Haro
Alexa Prize SocialBot Grand Challenge 4 (SGC4)
paper

Automatic Evaluation of Non-task Oriented Dialog Systems by Using Sentence Embeddings Projections and Their Dynamics
Mario Rodríguez-Cantelar, L.F. D'Haro and F. Matía
IWSDS 2020, Madrid, Spain, 2020.
paper youtube slides

An Assistant Robot Prototype with Social Capabilities
Gema Fernández-Blanco, Lucía García, Patricia Sánchez, Mario Rodríguez, Paloma de la Puente, Mª Guadalupe Sánchez-Escribano, Daniel Galán and Fernando Matía
Robocity 2030 Workshop, Madrid, Spain, 2019
Poster presentation
Driving Assistance System based on Detection of Driver Behaviour
Gema Fernández-Blanco, Álvaro Belloso, Mario Rodríguez, Paloma de la Puente, Mª Guadalupe Sánchez-Escribano, Daniel Galán and Fernando Matía
Robocity 2030 Workshop, Madrid, Spain, 2019
Poster presentation
Neural Model Applied to a Conversational Agent in a Closed Domain
Mario Rodríguez-Cantelar, Fernando Matía and Pablo San Segundo
XV CEA Intelligent Control Symposium
Contents EUROSIM 2019 Abstract Volume, pp. 51
paper abstracts-paper
Analysis of the Dialogue Management of a Generative Neuronal Conversational Agent
Mario Rodríguez-Cantelar
Master’s Thesis
paper

Analysis of the Neural Networks Architecture Applied to Conversational Agents
Mario Rodríguez-Cantelar
Bachelor’s Thesis
paper

ICT Natural Language
Topic: Develop an emotional guided Conversational Agent (chatbot), trained with Reinforcement Learning, capable of generating emotionally relevant and controlled responses while maintaining personality, internal emotions, and dialogue coherence.
Center: USC Institute for Creative Technologies
Place, Country: Los Angeles, California, U.S.
Date: September 1 – November 30, 2023

2020 Seventh Frederick Jelinek Memorial Summer Workshop
Topic: CHANEL: Chat/Dialogue Modeling and Evaluation – Development of multilingual and robust resources for open-domain dialogue evaluation, including automatic machine translation, sentiment analysis, and content moderation metadata. Focused on under-resourced languages and mitigating harmful or inappropriate training data through meta-annotation.
Center: Johns Hopkins University
Place, Country: Baltimore, Maryland, U.S.
Date: June 22 – August 7, 2020
GitHub youtube

CHANEL@JSALT2020: Extended Corpora Metadata using Microsoft Azure Cognitive Services
GitHub
AI Feedback Emotional Alignment
HuggingFace
DSTC11: Dialogue System Technology Challenge 11
HuggingFace

MENU