datasets
Training and evaluation data, with the modality, task and licence stated up front. Listed live from the Hugging Face Hub.
truthfulqa_va
TRUTHFULQA_VA Dataset
Dataset Summary
TruthfulQA_va is the Valencian version of the TruthfulQA dataset. This dataset is used to measure the truthfulness of a language model when generating answers to questions. It includes questions from different categories that some humans would answer wrongly due to false beliefs or misconceptions. Note that this version includes only the generation split.
Dataset Structure
Each row in the dataset includes the following… See the full description on the dataset page: https://huggingface.co/datasets/gplsi/truthfulqa_va.TerretaQA
🥘 TerretaQA Dataset
TerretaQA is a benchmark dataset containing 200 test questions designed for question answering about localities in the Valencian Community (Spain).
The dataset is intended to evaluate systems that require geographically grounded knowledge, cultural awareness, and regional understanding.
📌 Dataset Overview
Name: TerretaQA
Type: Question Answering (QA) benchmark
Domain: Local geography, culture, and administrative divisions of the Valencian… See the full description on the dataset page: https://huggingface.co/datasets/gplsi/TerretaQA.cieaCOVA
cieaCOVA Dataset
cieaCOVA is a Valencian-language (va) evaluation dataset designed to benchmark large language models (LLMs) on structured reasoning and generative tasks. The dataset contains 1,982 curated examples and is specifically developed for evaluation purposes — not for model training.
The dataset is organized into two task-oriented directories, each containing a train and test split:
multiple_choice/ (train.parquet, test.parquet) — multiple-choice question answering… See the full description on the dataset page: https://huggingface.co/datasets/gplsi/cieaCOVA.