InternScience/SFE

科学家首次考试（SFE）是一个旨在通过三个认知层次（科学信号感知、科学属性理解和科学比较推理）全面评估多模态大语言模型（MLLMs）科学认知能力的基准测试。数据集包含830个专家验证的视觉问答对（VQA pairs），涵盖五个高价值学科（天文学、化学、地球科学、生命科学和材料科学）的66个多模态任务。所有任务均为双语（英语和中文），以支持广泛的访问。这些任务不仅需要深入理解领域特定的知识和数据分析技能，还能显著提高研究效率并促进社会进步的进展。

The Scientists First Exam (SFE) benchmark is designed to comprehensively evaluate the scientific cognitive capabilities of Multimodal Large Language Models (MLLMs) through three cognitive levels: Scientific Signal Perception, Scientific Attribute Understanding, and Scientific Comparative Reasoning. SFE comprises 830 expert-verified Visual Question Answering (VQA) pairs across 66 multimodal tasks spanning five high-value disciplines: Astronomy, Chemistry, Earth, Life, and Materials Sciences. All tasks are bilingual (English & Chinese) to support broad accessibility. These tasks are designed not only to require a deep understanding of domain-specific knowledge and data analysis skills but also to significantly enhance research efficiency and facilitate advancements that benefit society.