CBCarlos Badenes-Olmedo

Carlos Badenes-Olmedo

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


Carlos Badenes-Olmedo has a total of 227 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    TailNLG: A Multilingual Benchmark Addressing Verbalization of Long-Tail Entities

    by Lia Draetta, Michael Oliverio, Virginia Ramón-Ferrer, Pier Felice Balestrucci, Flaviana Corallo, Carlos Badenes-Olmedo, Alessandro Mazzei, Marco Antonio Stranisci, Rossana Damiano

    Automatic verbalization of structured knowledge is a key task for making knowledge graphs accessible to non-expert users and supporting Retrieval-Augmented Generation systems. Although recent advances in Data-to-Text generation have improved multilingual coverage, little attention has been paid to potential biases in the verbalization of rare entities, frequently known as long-tail entities. In this work, we present the first systematic study of long-tail entities in Data-to-Text generation. We introduce TailNLG, a new multilingual benchmark in English, Italian, and Spanish, built from Wikidata and covering entities with varying levels of popularity. We evaluated three different families of large language models in zero-shot and few-shot settings and compared their performance on rare versus common entities, as well as against the established WebNLG benchmark. Our results suggest a tendency toward lower performance on long-tail entities: embedding-based scores are directionally lower for rare entities across most models, and within-model uncertainty as measured by Perplexity is consistently higher.

  • Open Access

    An unsupervised multidimensional data drift framework for measuring multilingual and cross-national political discourse shifts

    by Ibai Guillén-Pacho, Carlos Badenes-Olmedo, Oscar Corcho

    Political discourse on social networks continuously evolves across electoral cycles and communicative contexts, producing measurable changes in discourse distributions over time. This work introduces an unsupervised framework for detecting and quantifying these changes through multidimensional drift analysis. The approach models discourse evolution across semantic, lexical, thematic, and syntactic dimensions by combining complementary drift detection techniques over text, enabling systematic and reproducible analyses of discourse change. The framework is applied to multilingual political discourse published on party manifestos and Facebook between 2014 and 2023 across multiple electoral cycles and institutional levels in Spain and France. Results show that the proposed framework is able to detect, in a fully unsupervised manner, structural patterns consistent with known dynamics of political communication, including stable distinctions between campaign and routine discourse. At the same time, the framework provides measurable and objective indicators that enable more systematic analyses of discourse evolution across comparable political scenarios.

  • Open Access

    Spanish Triple-to-Text Benchmark on Low-Resource Large Language Models

    by Virginia Ramón-Ferrer, Carlos Badenes-Olmedo, Oscar Corcho

    The verbalization of structured data is a beneficial process for several applications. In the context of knowledge graphs (KGs), transforming Resource Description Framework (RDF) triples into natural language facilitates tasks such as KG documentation or alternative exploration methods for different user needs. While significant progress has been made on the English verbalization of KGs, Spanish remains an under-represented language for this task due to the lack of suitable resources. This hinders the development and evaluation of models capable of generating high-quality Spanish verbalizations. To tackle this problem, we create a Spanish adaptation of the WebNLG dataset, a benchmark consisting of over 45,000 verbalizations paired with DBpedia triple sets. To our knowledge, this is the first formal attempt to provide such a dataset in Spanish, which not only serves for data verbalization but can also potentially support the automated generation of RDF triples from text. We leverage this dataset to conduct a comprehensive evaluation of resource-efficient models for the Spanish triple-to-text task employing two different learning approaches: context learning (zero-shot, one-shot, and few-shot settings) and supervised learning through partial fine-tuning. Our results highlight the challenges of generating fluent and accurate Spanish text and demonstrate that partial fine-tuning of the evaluated models significantly improves performance.


  • Open Access

    ESQAD: A curriculum-aligned dataset for question answer generation in Spanish

    by Carlos Badenes-Olmedo, Paul Eyzaguirre-Barreda, Noa Chu-Artzt, Joaquin Gayoso-Cabada


  • Experiencias de Aprendizaje-Servicio en la UPM: 2025

    by Gabriel Lopez Font, Laura Recuero Pavón, Ana Rafaela Velasco Arranz, Mari Carmen Suárez-Figueroa, Carolina Chaya Romero, Paula Marina Blanco del Pozo, Isam Diab Lozano, Arminda Moreno Díaz, Carlos Badenes Olmedo, Marina Camarasa Rius, et al.

    El Aprendizaje-Servicio (ApS) es hoy una de las metodologías que mejor refleja el compromiso de la Universidad Politécnica de Madrid con una formación que va más allá de lo académico. A través de ella, estudiantado, profesorado (PDI) y personal técnico de gestión y de administración y servicios (PTGAS) tienen la oportunidad de aplicar sus conocimientos en contextos reales, contribuyendo al mismo tiempo a dar respuesta a necesidades de la sociedad. Desde su creación en 2019, la Oficina de Aprendizaje-Servicio de la UPM impulsa la incorporación de esta metodología en las distintas escuelas y centros de la universidad, promoviendo la colaboración con entidades sociales, administraciones públicas, asociaciones, fundaciones, centros educativos y organizaciones del tercer sector. Este trabajo ha permitido consolidar numerosas iniciativas que reflejan el potencial del ApS para conectar universidad y territorio. La convocatoria anual de proyectos de ApS constituye una de las principales herramientas para fomentar estas iniciativas. Los proyectos desarrollados durante el último curso han permitido abordar desafíos relacionados con la inclusión social, la sostenibilidad ambiental, la accesibilidad, la educación, la salud y otros ámbitos estrechamente vinculados con los Objetivos de Desarrollo Sostenible y la Agenda 2030. Esta tercera edición del e-book Experiencias de Aprendizaje-Servicio Universitario recoge algunos de estos proyectos, cuyos artículos se han organizado en torno a seis grandes ámbitos de actuación: la formación, la cultura y la concienciación; la sostenibilidad ambiental y la transición ecológica; el territorio, la ciudad y el patrimonio; la justicia social, la inclusión y la equidad; la innovación tecnológica con impacto social; y la salud, el bienestar y los entornos saludables. Esta agrupación permite ofrecer una visión estructurada de las principales líneas de trabajo en las que el ApS está contribuyendo a generar impacto desde la UPM. Cada una de las experiencias presentadas evidencia cómo el conocimiento generado en la universidad puede ponerse al servicio de la sociedad, favoreciendo al mismo tiempo un aprendizaje más significativo, comprometido y transformador para el estudiantado. Estos proyectos son el resultado del esfuerzo conjunto de profesorado, alumnado, PTGAS y entidades colaboradoras, y muestran la capacidad de la universidad para contribuir con soluciones innovadoras a problemas reales. Queremos expresar nuestro más sincero agradecimiento a todas las personas que han hecho posible las experiencias recogidas en esta publicación. Su dedicación, entusiasmo y vocación de servicio constituyen el mejor ejemplo de cómo la universidad puede contribuir activamente a la construcción de una sociedad más inclusiva, sostenible y justa. Confiamos en que esta nueva edición sirva para dar visibilidad a estas iniciativas, inspire a nuevos equipos docentes, estudiantes y personal de la universidad a incorporar el Aprendizaje-Servicio en sus actividades, y contribuya a seguir fortaleciendo esta comunidad comprometida con la innovación educativa y el impacto social.

  • Open Access

    MuHeQA: Zero-shot question answering over multiple and heterogeneous knowledge bases

    by Carlos Badenes-Olmedo, Oscar Corcho

    There are two main limitations in most of the existing Knowledge Graph Question Answering (KGQA) algorithms. First, the approaches depend heavily on the structure and cannot be easily adapted to other KGs. Second, the availability and amount of additional domain-specific data in structured or unstructured formats has also proven to be critical in many of these systems. Such dependencies limit the applicability of KGQA systems and make their adoption difficult. A novel algorithm is proposed, MuHeQA, that alleviates both limitations by retrieving the answer from textual content automatically generated from KGs instead of queries over them. This new approach (1) works on one or several KGs simultaneously, (2) does not require training data what makes it is domain-independent, (3) enables the combination of knowledge graphs with unstructured information sources to build the answer, and (4) reduces the dependency on the underlying schema since it does not navigate through structured content but only reads property values. MuHeQA extracts answers from textual summaries created by combining information related to the question from multiple knowledge bases, be them structured or not. Experiments over Wikidata and DBpedia show that our approach achieves comparable performance to other approaches in single-fact questions while being domain and KG independent. Results raise important questions for future work about how the textual content that can be created from knowledge graphs enables answer extraction.