![]() ![]() Answers with many tokens often resemble extractive snippets rather than canonical answers, so we discard answers with more than 5 tokens.Įvaluating on this diverse set of question-answer pairs is crucial, because all existing datasets have inherent biases that are problematic for open domain QA systems with learned retrieval. To gather an open version of this dataset, we only keep questions with short answers and discard the given evidence document. Natural Questions contains question from aggregated queries to Google Search (Kwiatkowski et al., 2019). Initial Data Collection and Normalization answer - List of possible answers to the question."City of Ekurhuleni Metropolitan Municipality" "Buffalo City Metropolitan Municipality", "City of Johannesburg Metropolitan Municipality", "City of Tshwane Metropolitan Municipality", "Nelson Mandela Bay Metropolitan Municipality", "question": "names of the metropolitan municipalities in south africa", The goal is to predict an English answer string for an input English question.Īll questions can be answered using the contents of English Wikipedia. Is an open domain question answering benchmark that is derived from Natural Questions. The NQ-Open task, introduced by Lee et.al. ![]()
0 Comments
Leave a Reply.AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |