2026-09-10T09:44:58Z https://drops.dagstuhl.de/oai

oai:drops-oai.dagstuhl.de:27548 2026-09-10T05:38:42Z ddc:004 open_access

RCC-8 as a Benchmark for Diagrammatic Reasoning in Multimodal Foundation Models Blackwell, Robert E Cohn, Anthony G Large Language Models Foundation Models Vision-Language Models Spatial Reasoning Diagrammatic Reasoning Diagrams are commonly used to define relations in qualitative spatial calculi, and humans routinely rely on diagrams for spatial reasoning. Thus the question arises: can multimodal AI foundation models make use of diagrams for qualitative spatial reasoning? Using the Region Connection Calculus (RCC-8), a well-known qualitative spatial calculus, we investigate three capabilities: whether Vision–Language Models (VLMs) can recognise RCC-8 spatial relation diagrams, whether text-to-image and text-to-video models can illustrate RCC-8 relations, and whether Large Language Models (LLMs) can exploit diagrammatic reasoning. We show that, while no state-of-the-art model is fully reliable, the best models typically recognise vector diagrams more accurately than raster diagrams, albeit with problems recognising relations with a precise tangent. Diagrams with squares or circles are more readily recognisable than triangles or blobs. State-of-the-art LLMs can answer composition questions remarkably well, but there is little evidence that they benefit from explicit prompting for diagrammatic reasoning. Although state-of-the-art image and video generation models are widely used to, for example, produce social media content, they struggle to provide accurate RCC-8 relation or conceptual neighbourhood illustrations except when SVG is the output format. Schloss Dagstuhl – Leibniz-Zentrum für Informatik Robert E Blackwell and Anthony G Cohn 2026 Is Part Of LIPIcs, Volume 393, 17th International Conference on Spatial Information Theory (COSIT 2026) InProceedings Text doc-type:ResearchArticle publishedVersion application/pdf doi:10.4230/LIPIcs.COSIT.2026.4 urn:nbn:de:0030-drops-275489 https://drops.dagstuhl.de/entities/document/10.4230/LIPIcs.COSIT.2026.4 eng https://creativecommons.org/licenses/by/4.0/legalcode