[2604.07956] MONETA: Multimodal Industry Classification through Geographic Information with Multi Agent Systems

Computer Science > Artificial Intelligence arXiv:2604.07956 (cs) [Submitted on 9 Apr 2026 (v1), last revised 10 Apr 2026 (this version, v2)] Title:MONETA: Multimodal Industry Classification through Geographic Information with Multi Agent Systems Authors:Arda Yüksel, Gabriel Thiem, Susanne Walter, Patrick Felka, Gabriela Alves Werb, Ivan Habernal View a PDF of the paper titled MONETA: Multimodal Industry Classification through Geographic Information with Multi Agent Systems, by Arda Y\"uksel and 5 other authors View PDF HTML (experimental) Abstract:Industry classification schemes are integral parts of public and corporate databases as they classify businesses based on economic activity. Due to the size of the company registers, manual annotation is costly, and fine-tuning models with every update in industry classification schemes requires significant data collection. We replicate the manual expert verification by using existing or easily retrievable multimodal resources for industry classification. We present MONETA, the first multimodal industry classification benchmark with text (Website, Wikipedia, Wikidata) and geospatial sources (OpenStreetMap and satellite imagery). Our dataset enlists 1,000 businesses in Europe with 20 economic activity labels according to EU guidelines (NACE). Our training-free baseline reaches 62.10% and 74.10% with open and closed-source Multimodal Large Language Models (MLLM). We observe an increase of up to 22.80% with the combination of multi-...