On November 18, 2025, at 10:00 AM CEST (9:00 AM PT), the webinar “AI-Driven Archaeology with LLMs — Detecting Archaeological Sites from Aerial Imagery” will take place, jointly organised by NCC Slovakia and NCC Portugal as part of the EuroCC project’s webinar series connecting AI, cultural heritage, and innovation.

The session will be led by Dr. Daniel Canedo, Research Fellow at the Institute of Electronics and Informatics Engineering of Aveiro (IEETA), University of Aveiro, Portugal. He will present real-world use cases where LLM-inspired models and Vision Transformers are applied to aerial and LiDAR imagery to detect hidden archaeological sites with high accuracy and minimal false positives.

Archaeological research is entering a new era powered by remote sensing and artificial intelligence. Sites like hillforts and burial mounds often have complex shapes, making them difficult to detect with traditional methods.

This session explores how multimodal learning (combining LiDAR with aerial imagery), Vision Transformers, and cross-modal attention mechanisms can significantly improve boundary detection and reduce false positives.

Using a human-in-the-loop process, where archaeologists validate and refine AI predictions, the model’s accuracy and interpretability are enhanced. In Northwest Iberia, this approach led to a 99.3% reduction in false positives after a single refinement cycle.

The webinar also covers predictive modelling techniques for identifying high-potential areas for future archaeological surveys — accelerating discovery while preserving cultural heritage.

This initiative reflects the EuroCC project’s commitment to bridging HPC and AI with societal and scientific challenges, showing how technology can support cultural innovation.

Registration here.