MME-Industry
收藏资源简介:
MME-Industry是由武汉人工智能研究院和中国科学院自动化研究所联合创建的多模态评估基准数据集,旨在评估多模态大语言模型(MLLMs)在工业环境中的表现。该数据集包含1050个问题-答案对,覆盖21个不同的工业领域,如电力、电子、纺织、钢铁和化工等。每个领域包含50个问题,所有问题均由领域专家手动创建和验证,确保了数据的完整性和专业性。数据集还提供了中英文版本,支持跨语言研究。数据集的应用领域主要集中在工业场景中的复杂问题解决,旨在评估模型在专业知识和推理能力方面的表现。
MME-Industry is a multimodal evaluation benchmark dataset jointly created by Wuhan AI Institute and the Institute of Automation, Chinese Academy of Sciences, aiming to evaluate the performance of multimodal large language models (MLLMs) in industrial environments. This dataset contains 1050 question-answer pairs, covering 21 distinct industrial sectors such as electric power, electronics, textile, iron and steel, chemical industry, etc. Each sector includes 50 questions, and all questions are manually created and verified by domain experts, ensuring the data's integrity and professionalism. The dataset also provides both Chinese and English versions, supporting cross-lingual research. Its application scenarios mainly focus on complex problem-solving in industrial settings, with the goal of evaluating the model's performance in terms of professional knowledge and reasoning abilities.

- 1MME-Industry: A Cross-Industry Multimodal Evaluation Benchmark武汉人工智能研究院, 中国科学院自动化研究所 · 2025年



