RetChemQA
收藏资源简介:
RetChemQA是一个专为网状化学领域设计的大型基准数据集,由加州大学伯克利分校的研究团队创建。该数据集包含约90,000个单跳和多跳问题与答案对,数据来源于约2,530篇研究论文,涵盖多个主流科学出版机构。数据集的创建过程利用了OpenAI的GPT-4 Turbo模型,该模型以其卓越的语言理解和生成能力著称。RetChemQA旨在为网状化学领域的机器学习算法开发和评估提供一个强有力的平台,特别适用于评估模型在处理复杂科学问题时的表现。
RetChemQA is a large-scale benchmark dataset specifically designed for the field of reticular chemistry, created by a research team at the University of California, Berkeley. This dataset contains approximately 90,000 single-hop and multi-hop question-answer pairs, sourced from around 2,530 research papers across multiple leading scientific publishing institutions. The dataset was constructed using OpenAI's GPT-4 Turbo model, which is renowned for its exceptional language understanding and generation capabilities. RetChemQA aims to provide a robust platform for the development and evaluation of machine learning algorithms in the field of reticular chemistry, and is particularly suitable for assessing model performance when handling complex scientific problems.

- 1Single and Multi-Hop Question-Answering Datasets for Reticular Chemistry with GPT-4-Turbo加州大学伯克利分校 · 2024年



