AnomalyR1：基于GRPO的端到端MLLM工业异常检测模型
AnomalyR1: A GRPO-based End-to-end MLLM for Industrial Anomaly Detection
- Nanjing University（南京大学）
机构由 AI 辅助整理，请以论文原文为准。
AI总结：
针对工业异常检测中缺陷样本稀缺、传统方法泛化性不足的问题，提出基于VLM-R1（MLLM）与ROAM增强GRPO的AnomalyR1端到端框架，在基准上优于现有方法，为少缺陷工业场景提供新方案。
AI中文摘要：
工业异常检测（IAD）因缺陷样本稀缺面临严峻挑战，亟需部署具备鲁棒泛化能力的模型以有效检测未知异常。传统方法常受限于手工特征或特定领域专家模型，难以解决这一局限，凸显范式转变的必要性。我们提出AnomalyR1，这一开创性框架利用VLM-R1（以卓越泛化性和可解释性著称的多模态大语言模型，MLLM）革新IAD。通过将MLLM与群体相对策略优化（GRPO）集成，并以我们新提出的推理结果对齐度量（ROAM）增强，AnomalyR1实现端到端解决方案，可自主处理图像和领域知识输入、通过分析推理并生成精确异常定位与掩码。基于最新多模态IAD基准，我们30亿参数的紧凑模型优于现有方法，建立了最先进结果。随着MLLM能力持续提升，本研究首次提出端到端VLM基IAD解决方案，展示ROAM增强GRPO的变革潜力，为缺陷数据有限的工业应用中下一代智能异常检测系统奠定前瞻性基石。
英文摘要：
Industrial Anomaly Detection (IAD) poses a formidable challenge due to the scarcity of defective samples, making it imperative to deploy models capable of robust generalization to detect unseen anomalies effectively. Traditional approaches, often constrained by hand-crafted features or domain-specific expert models, struggle to address this limitation, underscoring the need for a paradigm shift. We introduce AnomalyR1, a pioneering framework that leverages VLM-R1, a Multimodal Large Language Model (MLLM) renowned for its exceptional generalization and interpretability, to revolutionize IAD. By integrating MLLM with Group Relative Policy Optimization (GRPO), enhanced by our novel Reasoned Outcome Alignment Metric (ROAM), AnomalyR1 achieves a fully end-to-end solution that autonomously processes inputs of image and domain knowledge, reasons through analysis, and generates precise anomaly localizations and masks. Based on the latest multimodal IAD benchmark, our compact 3-billion-parameter model outperforms existing methods, establishing state-of-the-art results. As MLLM capabilities continue to advance, this study is the first to deliver an end-to-end VLM-based IAD solution that demonstrates the transformative potential of ROAM-enhanced GRPO, positioning our framework as a forward-looking cornerstone for next-generation intelligent anomaly detection systems in industrial applications with limited defective data.
