Golden Touchstone
收藏资源简介:
Golden Touchstone是由IDEA研究院和香港科技大学(广州)等机构联合创建的综合性双语金融大语言模型基准数据集。该数据集涵盖了22个高质量数据集,跨越中英文两种语言,涉及八个核心金融NLP任务。数据集的创建过程结合了广泛的开源数据收集和行业特定需求,旨在全面评估模型的语言理解和生成能力。Golden Touchstone的应用领域主要集中在金融领域,旨在解决现有金融基准数据集在语言和任务覆盖上的不足,提供一个全面评估金融大语言模型的工具。
Golden Touchstone is a comprehensive bilingual financial large language model benchmark dataset jointly created by institutions including IDEA Research Institute and The Hong Kong University of Science and Technology (Guangzhou), as well as other relevant organizations. This dataset includes 22 high-quality datasets, spans both Chinese and English languages, and covers eight core financial natural language processing (NLP) tasks. The development of the dataset combines extensive open-source data collection and industry-specific requirements, aiming to comprehensively evaluate the language understanding and generation capabilities of models. Golden Touchstone is mainly applied in the financial field, and it is designed to address the gaps in language and task coverage of existing financial benchmark datasets, providing a comprehensive tool for evaluating financial large language models.

- 1Golden Touchstone: A Comprehensive Bilingual Benchmark for Evaluating Financial Large Language ModelsIDEA研究院, 香港科技大学(广州), 香港科技大学, 南京大学, 华南师范大学 · 2024年



