Y-NQ: English-Yorùbá Evaluation dataset for Open-Book Reading Comprehension with Open-Ended Questions

The purpose of this work is to share an English-Yorb evaluation dataset for openbook reading comprehension with open-ended questions to assess the performance of models both in a high-and a low-resource language.The dataset contains 358 questions and answers on 338 English documents and 208 Yorb documents.Experiments show a consistent disparity in performance between the two languages, with Yorb falling behind English for automatic metrics even if documents are much shorter for this language.For a small set of documents with comparable length, performance of Yorb drops by 2.5 times and this comparison is validated with human evaluation.When analyzing performance by length, we observe that Yorb decreases performance dramatically for documents that reach 1500 words while English performance is barely affected at that length.Our dataset opens the door to showcasing if English LLM reading comprehension capabilities extend to Yorb, which for the evaluated LLMs is not the case.