2025 Proceedings

Large language models (LLMs) and foundation models are revolutionizing GeoAI as they need little or no fine-tuning. Join us to discover how models such as Segment Anything Model (SAM) and Prithvi can be used to extract features from geospatial imagery. Learn how LLMs like Generative Pre-trained Transformer (GPT) can be prompted to extract geospatial insights from unstructured data. Presenters also show how Python developers can use advanced techniques like low-rank adaptation (LoRA) to fine-tune these models and integrate them with libraries like LangChain to build GeoAI agents and assistants.