Automatic vision inspection holds significant importance in industry inspection. While multimodal large language models (MLLMs) exhibit strong language understanding capabilities and hold promise for this task, their performance remains significantly inferior to that of human experts. In this context, we identify two key challenges: (i) insufficient integration of anomaly detection (AD) knowledge during pre-training, and (ii) the lack of technically precise and context-aware language generation for anomaly reasoning. To address these issues, we propose ADSeeker, an anomaly task assistant designed to enhance inspection performance through knowledge-grounded reasoning. ADSeeker first leverages a curated visual document knowledge base, SEEK-M&V, which we construct to address the limitations of existing resources that rely solely on unstructured text. SEEK-M\&V includes semantic-rich descriptions and image-document pairs, enabling more comprehensive anomaly understanding. To effectively retrieve and utilize this knowledge, we introduce the Query Image-Knowledge Retrieval-Augmented Generation Q2K RAG framework. To further enhance the performance in zero-shot anomaly detection (ZSAD), ADSeeker leverages the Hierarchical Sparse Prompt mechanism and type-level features to efficiently extract anomaly patterns. Furthermore, to tackle the challenge of limited industry anomaly detection (IAD) data, we introduce the largest-scale AD dataset, Multi-type Anomaly MulA, encompassing 72 multi-scale defect types across 26 categories. Extensive experiments show that our plug-and-play framework, ADSeeker, achieves state-of-the-art zero-shot performance on several benchmark datasets.

翻译：自动视觉检测在工业检测中具有重要意义。尽管多模态大语言模型（MLLMs）展现出强大的语言理解能力并有望胜任此任务，但其性能仍显著落后于人类专家。在此背景下，我们识别出两个关键挑战：（i）预训练阶段异常检测（AD）知识整合不足，以及（ii）缺乏技术精确且上下文感知的语言生成能力用于异常推理。为解决这些问题，我们提出ADSeeker，一种通过知识驱动推理增强检测性能的异常任务助手。ADSeeker首先利用我们构建的精选视觉文档知识库SEEK-M&V，以弥补现有资源仅依赖非结构化文本的局限性。SEEK-M&V包含语义丰富的描述和图像-文档对，能够实现更全面的异常理解。为有效检索和利用该知识，我们引入Query图像-知识检索增强生成框架Q2K RAG。为进一步提升零样本异常检测（ZSAD）性能，ADSeeker采用层次化稀疏提示机制和类型级特征高效提取异常模式。此外，针对工业异常检测（IAD）数据稀缺的难题，我们构建了最大规模的多类型异常数据集MulA，涵盖26个类别的72种多尺度缺陷类型。大量实验表明，我们的即插即用框架ADSeeker在多个基准数据集上实现了最先进的零样本检测性能。