Iria de-Dios-Flores — accepted AI-conference papers

2025

IberoBench: A Benchmark for LLM Evaluation in Iberian Languages

Irene Baucells, Javier Aula-Blasco, Iria de-Dios-Flores, Silvia Paniagua Suárez, Naiara Perez, Anna Salles +9

COLING 2025main

The current best practice to measure the performance of base Large Language Models is to establish a multi-task benchmark that covers a range of capabilities of interest. Currently, however, such benchmarks are only available in a few high-resource languages. To address this situation, we present Ib…

Cited by 2 Source ↗PDF ↗Scholar ↗

2025

Truth Knows No Language: Evaluating Truthfulness Beyond English

Blanca Calvo Figueras, Eneko Sagarzazu, Julen Etxaniz, Jeremy Barnes, Pablo Gamallo, Iria de-Dios-Flores +1

ACL 2025long

We introduce a professionally translated extension of the TruthfulQA benchmark designed to evaluate truthfulness in Basque, Catalan, Galician, and Spanish. Truthfulness evaluations of large language models (LLMs) have primarily been focused on English. However, the ability of LLMs to maintain truthf…

Cited by 0 Source ↗PDF ↗code ↗Scholar ↗

2023

Dependency resolution at the syntax-semantics interface: psycholinguistic and computational insights on control dependencies

Iria de-Dios-Flores, Juan Garcia Amboage, Marcos Garcia

ACL 2023long

Using psycholinguistic and computational experiments we compare the ability of humans and several pre-trained masked language models to correctly identify control dependencies in Spanish sentences such as ‘José le prometió/ordenó a María ser ordenado/a’ (‘Joseph promised/ordered Mary to be tidy’). T…

Cited by 5 Source ↗PDF ↗code ↗Scholar ↗