Academic profile
Academic titles are retained in their original language.
53 results
UPM Match uses language and knowledge technologies to support tutor selection and academic collaboration through thematic profiles and explainable recommendations.
A service-learning experience combining Easy-to-Read methods and AI to support reading comprehension in primary education.
Adaptation of OntoPortal for the TeresIA terminology metasearch service, integrating OntoLex lexical entries, forms and senses for multilingual search and reuse.
A Spanish curriculum-aligned question-answer dataset combining official examination questions and LLM-generated material. The contribution covers dataset design, cognitive criteria and educational evaluation.
MIRA supports the post-marking workflow for paper multiple-choice examinations, including optical mark recognition, review of scanned examinations and structured handling of student appeals.
NEXA-ACE combines functional test suites and language-model support to evaluate programming assignments, including basic, complex and edge cases.
Evaluation of open language models for syntactic adaptation to Easy-to-Read guidelines, using manually annotated sentences and tasks involving appositions and relative clauses.
An institutional study of question generation and evaluation from academic documents. The contribution covers the methodology, implementation and quantitative and qualitative analysis of the pilot.
A Spanish adaptation of WebNLG for generating text from RDF triples, evaluating efficient language models with in-context learning and partial fine-tuning.
Dynamic topic models and unsupervised labelling for exploring how concepts evolve in coronavirus scientific literature.
MeriCOIN studies physical, 3D-printed tokens as rewards within gamified activities in Software Engineering Fundamentals.
An ingestion and enrichment architecture for EU Contract Hub, supporting the exploration and analysis of European healthcare procurement within ProCure.
A technical report on opportunities, limitations and educational challenges in integrating artificial intelligence tools into higher education.
An experimental study of language models acting as domain experts in validation tasks, using questions with controlled and potentially biased contexts.
A schema-independent question-answering approach combining heterogeneous knowledge bases through natural-language verbalisation of triples.
The FLARES shared task studies linguistic reliability in Spanish news, including 5W1H extraction and reliability classification. The contribution focuses on information extraction and evaluation design.
PODIO represents political discourse, proposals, ideologies, actors, audiences and temporal context using ontology-based modelling.
Conversational language models generate interpretable labels for probabilistic topics from representative terms. The contribution includes scientific supervision, experimental design and analysis.