IAD-R1：在工业异常检测中强化一致性推理
IAD-R1: Reinforcing Consistent Reasoning in Industrial Anomaly Detection
AI总结：
针对工业异常检测中缺陷样本稀缺和VLMs性能受限的问题，提出通用后训练框架IAD-R1，通过PA-SFT和SC-GRPO两阶段策略增强模型推理与解释能力，使0.5B模型在零样本下超越GPT-4.1等商业大模型。
摘要：
工业异常检测是现代制造的关键组件，但缺陷样本的稀缺性将传统检测方法限制在特定场景应用。尽管视觉语言模型（VLMs）在泛化能力上展现出显著优势，其在工业异常检测中的性能依然有限。为应对该挑战，我们提出了IAD-R1，这是一个适用于不同架构和参数规模VLMs的通用后训练框架，能大幅增强其异常检测能力。IAD-R1采用两阶段训练策略：感知激活监督微调（PA-SFT）阶段利用精心构建的高质量思维链数据集（Expert-AD）进行训练，增强异常感知能力并建立推理到答案的关联；结构化对照组相对策略优化（SC-GRPO）阶段采用精心设计的奖励函数，实现从“异常感知”到“异常解释”的能力飞跃。实验结果表明，IAD-R1在7个VLMs上均取得显著提升，在DAGM数据集上提升最大，平均准确率比0.5B基线高43.3%。值得注意的是，使用IAD-R1训练的0.5B参数模型在零样本设置下超越了包括GPT-4.1和Claude-Sonnet-4在内的商业模型，证明了IAD-R1的有效性与优越性。数据集、代码及所有模型权重将在 https://github.com/Yanhui-Lee/IAD-R1 公开。
英文摘要：
Industrial anomaly detection is a critical component of modern manufacturing, yet the scarcity of defective samples restricts traditional detection methods to scenario-specific applications. Although Vision-Language Models (VLMs) demonstrate significant advantages in generalization capabilities, their performance in industrial anomaly detection remains limited. To address this challenge, we propose IAD-R1, a universal post-training framework applicable to VLMs of different architectures and parameter scales, which substantially enhances their anomaly detection capabilities. IAD-R1 employs a two-stage training strategy: the Perception Activation Supervised Fine-Tuning (PA-SFT) stage utilizes a meticulously constructed high-quality Chain-of-Thought dataset (Expert-AD) for training, enhancing anomaly perception capabilities and establishing reasoning-to-answer correlations; the Structured Control Group Relative Policy Optimization (SC-GRPO) stage employs carefully designed reward functions to achieve a capability leap from "Anomaly Perception" to "Anomaly Interpretation". Experimental results demonstrate that IAD-R1 achieves significant improvements across 7 VLMs, the largest improvement was on the DAGM dataset, with average accuracy 43.3% higher than the 0.5B baseline. Notably, the 0.5B parameter model trained with IAD-R1 surpasses commercial models including GPT-4.1 and Claude-Sonnet-4 in zero-shot settings, demonstrating the effectiveness and superiority of IAD-R1. The dataset, code, and all model weights will be publicly available at https://github.com/Yanhui-Lee/IAD-R1.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
