多模态模型能否正确解析天文视觉问答数据集？
原标题：Star Light, Star Bright: Can Multimodal Models Parse an Astronomy Visual Question Answering Dataset Right?
AI 摘要
研究者构建了一个面向天文天空图像的领域特定VQA数据集，包含2001张部分合成图像（等分为等高线图、NASA SkyView真实望远镜图像和GMM生成的假天空图像）及47357个问题，每张图同时提供原始版本和模拟历史页面老化的退化版本。测试三种低成本专有LMM后发现，模型在高精度回答领域特定问题时经常失败；与104张经人工验证的历史期刊图像对比，角度范围中位误差相近。代码和数据集已分别在GitHub和HuggingFace开源。
正文节选
Star Light, Star Bright: Can Multimodal Models Parse an Astronomy Visual Question Answering Dataset Right? Abstract Figures are essential for scientific communication, and the field of Visual Question Answering (VQA) aims to test the figure understanding capabilities of large multimodal models (LMMs). Here, we investigate the accuracy of several low-cost LMMs on domain-specific VQA tasks for astronomical images of the sky. As instrument for this task we build a domain-specific VQA dataset compri