HiTZ/truthful_judge

TruthfulQA-Multi评判者训练数据是一个用于训练大型语言模型（LLM）评判者模型的数据集，旨在评估其他语言模型生成文本的真实性。该数据集包含英语以及西班牙语、加泰罗尼亚语、加利西亚语和巴斯克语等多种语言的训练实例，适用于细化和评估语言模型在真实性判断方面的性能。

The TruthfulQA-Multi Judge Training Data is a dataset designed for fine-tuning large language models (LLM) to act as judge models, aiming to evaluate the truthfulness of text generated by other language models. It includes training instances in English as well as in Spanish, Catalan, Galician, and Basque languages, suitable for refining and assessing the performance of language models in truthfulness judgment.