MONETA: Multimodal Industry Classification through Geographic Information with Multi Agent Systems

Arda Yüksel, Gabriel Thiem, Susanne Walter, Patrick Felka, Gabriela Alves Werb, Ivan Habernal, Trustworthy Human Language Technologies Technical University of Darmstadt, Germany · 2026

Read MONETA: Multimodal Industry Classification through Geographic Information with Multi Agent Systems by Arda Yüksel, Gabriel Thiem, Susanne Walter, Patrick Felka, Gabriela Alves Werb, Ivan Habernal, Trustworthy Human Language Technologies Technical University of Darmstadt, Germany (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2604-07956.txt.

Description

Yüksel and five colleagues present MONETA, the first multimodal benchmark for industry classification with geographic information. These schemes anchor public and company databases, but manual labeling of huge registers costs dearly, and retraining models for every update demands fresh data. MONETA replicates expert checks from existing mixed resources, pairing text from web pages, Wikipedia, and Wikidata with OpenStreetMap and orbital photos. Its dataset covers 1,000 European firms across 20 activity tags under EU NACE rules. A training-free baseline scores 62.10 percent with open models and about 74 percent with closed models.

Catalog record

CREATORArda Yüksel, Gabriel Thiem, Susanne Walter, Patrick Felka, Gabriela Alves Werb, Ivan Habernal, Trustworthy Human Language Technologies Technical University of Darmstadt, Germany
YEAR2026
COLLECTIONPAPERS
FORMATTXT
FILE SIZE66640
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2604.07956

Browse the full library · About the Tower of Babel