Quantifying and extending the coverage of spatial categorization data sets

2026-03-10

AI Summary

Existing spatial categorization datasets, such as TRPS, exhibit limitations in scene diversity and linguistic coverage, hindering a comprehensive characterization of cross-linguistic variation in spatial relation expressions. This work proposes a novel approach that leverages large language models (LLMs) to guide dataset expansion by generating spatial relation labels aligned with human annotations, thereby informing the selection of new scenes and languages. Through this method, the TRPS dataset is systematically extended with 42 additional scenes, achieving superior spatial coverage compared to its two prior versions. The resulting enhanced dataset lays a foundational framework for constructing a large-scale cross-linguistic spatial categorization resource encompassing dozens of languages and hundreds of scenes.

Technology Category

• Knowledge Representation and Reasoning: Geometric, Spatial, and Temporal Reasoning
• Natural Language Processing: Language Grounding & Multi-modal NLP
• Computer Vision: Language and Vision

Application Category

• Search and Retrieval-Augmented AI: Multilingual and cross-lingual Web search
• Web Mining and Content Analysis: Mining multimedia, multimodal, multilingual, cross-lingual Web data
• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions

Abstract

Variation in spatial categorization across languages is often studied by eliciting human labels for the relations depicted in a set of scenes known as the Topological Relations Picture Series (TRPS). We demonstrate that labels generated by large language models (LLMs) align relatively well with human labels, and show how LLM-generated labels can help to decide which scenes and languages to add to existing spatial data sets. To illustrate our approach we extend the TRPS by adding 42 new scenes, and show that this extension achieves better coverage of the space of possible scenes than two previous extensions of the TRPS. Our results provide a foundation for scaling towards spatial data sets with dozens of languages and hundreds of scenes.

Problem

• spatial categorization
• Topological Relations Picture Series
• cross-linguistic variation
• dataset coverage
• scene extension

Innovation

• large language models
• spatial categorization
• Topological Relations Picture Series
• dataset extension
• cross-linguistic variation