GeoDANO：具有领域无关视觉编码器的几何视觉语言模型
GeoDANO: Geometric VLM with Domain Agnostic Vision Encoder
AI总结：
GeoDANO提出一种领域无关视觉编码器，通过GeoCLIP和领域适应策略，提升几何特征识别，在平面几何问题解决上超越现有方法和GPT-4o。
摘要：
我们提出了GeoDANO，一种具有领域无关视觉编码器的几何视觉语言模型（VLM），用于解决平面几何问题。尽管VLM已被用于解决几何问题，但它们识别几何特征的能力仍未得到充分分析。为了弥补这一空白，我们提出了一个基准，用于评估视觉几何特征的识别，包括点、线等基本元素，以及正交性等关系。我们的初步研究表明，通用VLM中常用的视觉编码器（如OpenCLIP）无法检测这些特征，并且难以跨领域泛化。为了克服这一局限，我们开发了GeoCLIP，一种基于CLIP的模型，在合成的几何图形-标题对上训练。基准结果表明，GeoCLIP在识别几何特征方面优于现有的视觉编码器。然后，我们提出了我们的VLM GeoDANO，它通过一种针对未见图表风格的领域适应策略增强了GeoCLIP。GeoDANO在MathVerse上优于专门用于平面几何问题的方法和GPT-4o。实现可在https://github.com/ml-postech/GeoDANO获取。
英文摘要：
We introduce GeoDANO, a geometric vision-language model (VLM) with a domain-agnostic vision encoder, for solving plane geometry problems. Although VLMs have been employed for solving geometry problems, their ability to recognize geometric features remains insufficiently analyzed. To address this gap, we propose a benchmark that evaluates the recognition of visual geometric features, including primitives such as dots and lines, and relations such as orthogonality. Our preliminary study shows that vision encoders often used in general-purpose VLMs, e.g., OpenCLIP, fail to detect these features and struggle to generalize across domains. To overcome the limitation, we develop GeoCLIP, a CLIP-based model trained on synthetic geometric diagram--caption pairs. Benchmark results show that GeoCLIP outperforms existing vision encoders in recognizing geometric features. We then propose our VLM, GeoDANO, which augments GeoCLIP with a domain adaptation strategy for unseen diagram styles. GeoDANO outperforms specialized methods for plane geometry problems and GPT-4o on MathVerse. The implementation is available at https://github.com/ml-postech/GeoDANO.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
