arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2610.16769cs.AI
收藏

地理空间矢量几何作为视觉语言模型的原生空间模态

Geospatial Vector Geometry as a Native Spatial Modality for Vision-Language Models

The Hong Kong University of Science and Technology (Guangzhou)(香港科技大学(广州));The Hong Kong University of Science and Technology(香港科技大学)

Huang Fei, Boyi Chen, Hao Xue, Yang Yue, Yehong Xu, Xiaofang Zhou

AI总结:

本文提出将地理空间矢量几何作为视觉语言模型的原生空间模态,构建GeoVec-VLM模型,在5项空间推理任务中平均准确率较最强基线提升约11%,为VLM空间推理提供了新的空间基础。

摘要:

地理空间推理将视觉语言模型(VLM)与现实世界的行动和决策联系起来,可靠的推理需要准确的空间信息和有效的空间表示。然而,当前的VLM主要从语言和视觉中获取这些信息;在更大的地理尺度下,有限的视觉观测使得难以建立准确的空间参考并恢复精确的空间关系,这构成了VLM空间感知与推理的瓶颈。本文提出地理空间矢量几何可成为VLM内的原生空间模态,并引入GeoVec-VLM模型:首先,开发保留矢量对象空间结构与排列的结构化几何表示,构建显式空间基础;其次,引入几何交互策略,将该基础融入多模态计算,使其根据查询的空间关系与视觉内容、语言参考进行交互。这些设计共同使感知与推理能基于共享几何参考运行。在5项空间推理任务中,GeoVec-VLM相比最强基线的平均准确率提升约11%,在地理迁移场景下仍保持该提升,且表现出更强的几何一致性。研究结果表明,地理空间矢量几何为VLM的空间推理提供了有前景的内部空间基础。

英文摘要:

Geospatial reasoning links VLMs to real-world action and decision making. Reliable reasoning requires accurate spatial information and effective spatial representations. However, current VLMs obtain this information mainly from language and vision. At larger geographic scales, limited visual observations make it difficult to establish accurate spatial references and recover precise spatial relations. This leads to a bottleneck for VLM spatial perception and reasoning. This paper proposes that geospatial vector geometry can become a native spatial modality within a VLM and introduces GeoVec-VLM. First, we develop a structured geometry representation that preserves the spatial structure and arrangement of vector objects, building an explicit spatial basis. Second, we introduce a geometry interaction strategy that brings this basis into multimodal computation, where it interacts with visual content and language references according to the queried spatial relations. Together, these designs allow perception and reasoning to operate on shared geometric references. Across five spatial reasoning tasks, GeoVec-VLM improves mean accuracy by about 11\% relative to the strongest baseline, maintains its gains under geographic transfer, and shows greater geometric consistency. Our results show that geospatial vector geometry offers a promising internal spatial basis for VLM spatial reasoning. https://github.com/TXFR/GeoVector-VLM

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。

↑