{"ID":2861931,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2510.00603","arxiv_id":"2510.00603","title":"LVLMs as inspectors: an agentic framework for category-level structural defect annotation","abstract":"Automated structural defect annotation is essential for ensuring infrastructure safety while minimizing the high costs and inefficiencies of manual labeling. A novel agentic annotation framework, Agent-based Defect Pattern Tagger (ADPT), is introduced that integrates Large Vision-Language Models (LVLMs) with a semantic pattern matching module and an iterative self-questioning refinement mechanism. By leveraging optimized domain-specific prompting and a recursive verification process, ADPT transforms raw visual data into high-quality, semantically labeled defect datasets without any manual supervision. Experimental results demonstrate that ADPT achieves up to 98% accuracy in distinguishing defective from non-defective images, and 85%-98% annotation accuracy across four defect categories under class-balanced settings, with 80%-92% accuracy on class-imbalanced datasets. The framework offers a scalable and cost-effective solution for high-fidelity dataset construction, providing strong support for downstream tasks such as transfer learning and domain adaptation in structural damage assessment.","short_abstract":"Automated structural defect annotation is essential for ensuring infrastructure safety while minimizing the high costs and inefficiencies of manual labeling. A novel agentic annotation framework, Agent-based Defect Pattern Tagger (ADPT), is introduced that integrates Large Vision-Language Models (LVLMs) with a semantic...","url_abs":"https://arxiv.org/abs/2510.00603","url_pdf":"https://arxiv.org/pdf/2510.00603v1","authors":"[\"Sheng Jiang\",\"Yuanmin Ning\",\"Bingxi Huang\",\"Peiyin Chen\",\"Zhaohui Chen\"]","published":"2025-10-01T07:31:42Z","proceeding":"cs.CV","tasks":"[\"cs.CV\"]","methods":"[\"Language Model\"]","has_code":false}