PRESTA 2025: Preguntas y Respuestas sobre Tablas en Español

The objective of this task is to develop Question Answering systems capable of answering questions posed over tabular data in Spanish using the DataBenchSPA benchmark. It aims to evaluate the ability of models to interpret real tables with multiple rows and columns and generate accurate answers of different types (boolean, categorical, numerical, or lists).

Publication
Osés Grijalba, J. et al. 2025. Overview of PRESTA at IberLEF 2025: Question Answering Over Tabular Data In Spanish. Procesamiento del Lenguaje Natural, 75, pp. 475-486.
Competition
PRESTA: Preguntas y Respuestas sobre Tablas en Español
Language
Spanish
URL Task
https://www.codabench.org/competitions/5538/
NLP topic
question answering
Dataset
Spa-DataBench
Year
2025
Publication link
http://journal.sepln.org/sepl…
Ranking metric
Accuracy

Task results

System Accuracy
ITU NLP 0.8700
sonrobok4 0.8700
MRT 0.8500
LyS 0.7800
quang3010 0.7500

If you have published a result better than those on the list, send a message to odesia-comunicacion@lsi.uned.es indicating the result and the DOI of the article, along with a copy of it if it is not published openly.