Low-Surface Brightness Galaxies and Artifacts

该数据集通过视觉-语言模型实现了对低表面亮度星系和人工制品的零样本分类。它展示了视觉-语言模型在不重新训练的情况下，在分类任务中的潜力，同时强调了与定制训练模型在准确性方面的比较。该数据集的任务是对天文图像进行零样本分类。This dataset achieves zero-shot classification of low surface brightness galaxies and artifacts via vision-language models (VLMs). It showcases the potential of vision-language models in classification tasks without the need for retraining, while emphasizing the accuracy comparison against custom-trained models. The core task of this dataset is zero-shot classification for astronomical images.