Gespeichert in:
| Hauptverfasser: | , , , , , , , , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2025
|
| Schlagworte: | |
| Online-Zugang: | https://arxiv.org/abs/2501.16254 |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866915124103610368 |
|---|---|
| author | Lee, Chaehong Paramanayakam, Varatheepan Karatzas, Andreas Jian, Yanan Fore, Michael Liao, Heming Yu, Fuxun Li, Ruopu Anagnostopoulos, Iraklis Stamoulis, Dimitrios |
| author_facet | Lee, Chaehong Paramanayakam, Varatheepan Karatzas, Andreas Jian, Yanan Fore, Michael Liao, Heming Yu, Fuxun Li, Ruopu Anagnostopoulos, Iraklis Stamoulis, Dimitrios |
| contents | We present GeoLLM-Squad, a geospatial Copilot that introduces the novel multi-agent paradigm to remote sensing (RS) workflows. Unlike existing single-agent approaches that rely on monolithic large language models (LLM), GeoLLM-Squad separates agentic orchestration from geospatial task-solving, by delegating RS tasks to specialized sub-agents. Built on the open-source AutoGen and GeoLLM-Engine frameworks, our work enables the modular integration of diverse applications, spanning urban monitoring, forestry protection, climate analysis, and agriculture studies. Our results demonstrate that while single-agent systems struggle to scale with increasing RS task complexity, GeoLLM-Squad maintains robust performance, achieving a 17% improvement in agentic correctness over state-of-the-art baselines. Our findings highlight the potential of multi-agent AI in advancing RS workflows. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2501_16254 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Multi-Agent Geospatial Copilots for Remote Sensing Workflows Lee, Chaehong Paramanayakam, Varatheepan Karatzas, Andreas Jian, Yanan Fore, Michael Liao, Heming Yu, Fuxun Li, Ruopu Anagnostopoulos, Iraklis Stamoulis, Dimitrios Machine Learning We present GeoLLM-Squad, a geospatial Copilot that introduces the novel multi-agent paradigm to remote sensing (RS) workflows. Unlike existing single-agent approaches that rely on monolithic large language models (LLM), GeoLLM-Squad separates agentic orchestration from geospatial task-solving, by delegating RS tasks to specialized sub-agents. Built on the open-source AutoGen and GeoLLM-Engine frameworks, our work enables the modular integration of diverse applications, spanning urban monitoring, forestry protection, climate analysis, and agriculture studies. Our results demonstrate that while single-agent systems struggle to scale with increasing RS task complexity, GeoLLM-Squad maintains robust performance, achieving a 17% improvement in agentic correctness over state-of-the-art baselines. Our findings highlight the potential of multi-agent AI in advancing RS workflows. |
| title | Multi-Agent Geospatial Copilots for Remote Sensing Workflows |
| topic | Machine Learning |
| url | https://arxiv.org/abs/2501.16254 |