FRIEDA: 500 Multimodal Examples for Open-Ended Cartographic Reasoning

FRIEDA consists of 500 multimodal examples for open-ended cartographic reasoning, developed by knowledge-computing and released in late 2025. The benchmark pairs real-world map images with natural-language questions and reference answers to evaluate spatial reasoning capabilities.

Use Cases

• Visual Question Answering (VQA) focused on topological and metric GIS relations
• Evaluating multi-step reasoning using the natural-language question and reference answer fields
• Testing cross-map grounding by analyzing reference and contextual map images

Strengths

• 500 curated examples
• Targets specific GIS relation types including topological, metric, and directional
• Requires multi-step reasoning and cross-map grounding

Limitations

• Small sample size of 500 records
• Domain-specific focus on cartography limits general visual reasoning evaluation

Provenance

Source

knowledge-computing (associated with Arxiv 2512.08016)

Freshness

Last updated January 2026