{"ID":2886696,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2508.02073","arxiv_id":"2508.02073","title":"Large model retrieval enhancement framework for construction site risk identification","abstract":"This study addresses construction site hazard identification by proposing a retrieval-augmented framework that enhances large language models (LLMs) without requiring fine-tuning. Current LLM-based approaches face limitations: image-text matching struggles with complex hazards, while instruction tuning lacks generalization and is resource-intensive. Our method dynamically integrates external knowledge and retrieved similar cases via prompt tuning, overcoming LLMs' limitations in domain knowledge and feature correlation. The framework comprises a case database, an image retrieval module, and an LLM-based reasoning module. Evaluated on real-site data, our approach boosted GLM-4V's accuracy to 50%, a 35.49% improvement over baselines, with consistent gains across hazard types. Ablation studies validated the effectiveness of our image retrieval strategy, showing the superiority of our LPIPS- and CLIP-based method. The proposed technique significantly improves identification accuracy and contextual understanding, demonstrating strong generalization and offering a practical path for intelligent safety risk detection in construction.","short_abstract":"This study addresses construction site hazard identification by proposing a retrieval-augmented framework that enhances large language models (LLMs) without requiring fine-tuning. Current LLM-based approaches face limitations: image-text matching struggles with complex hazards, while instruction tuning lacks generaliza...","url_abs":"https://arxiv.org/abs/2508.02073","url_pdf":"https://arxiv.org/pdf/2508.02073v2","authors":"[\"Jiawei Li\",\"Chengye Yang\",\"Yaochen Zhang\",\"Weilin Sun\",\"Lei Meng\",\"Xiangxu Meng\"]","published":"2025-08-04T05:28:58Z","proceeding":"cs.AI","tasks":"[\"cs.AI\"]","methods":"[\"Large Language Model\",\"Language Model\"]","has_code":false}