This research introduces SDIGLM, a novel approach that combines large language models with multi-modal analysis to identify and describe structural damage in civil engineering applications.

• Enables comprehensive damage analysis beyond simple classification
• Provides detailed textual descriptions of structural problems
• Outperforms traditional computer vision models through multi-modal reasoning
• Offers practical applications for infrastructure maintenance and safety

This innovation matters because it addresses critical limitations in current structural assessment methods, potentially transforming how engineers monitor and maintain infrastructure by combining visual analysis with sophisticated language understanding.

SDIGLM: Leveraging Large Language Models and Multi-Modal Chain of Thought for Structural Damage Identification