InspectorGPT: A Comparative Reasoning Enhanced VLM for Comprehensive Industrial Anomaly Detection

2026-08-30

AI Summary

论文提出InspectorGPT，一种基于对比推理的视觉-语言模型，用于工业异常检测，通过与无缺陷参考图像比较，提高对未知异常的识别能力。

Abstract

Industrial anomaly detection is a critical component of modern manufacturing. Most traditional unsupervised methods rely on modelling normal feature distributions, inherently limiting generalization to unknown categories. To improve generalizability, some recent methods incorporate vision-language models (VLMs) for zero-shot detection via text prompts. However, we observe that reasoning-oriented post-training can cause anomaly discrimination to collapse, with some fine-tuned models performing worse than their base VLMs. Existing methods also provide only textual decisions or coarse boxes, without pixel-level segmentation. A more explicit detection principle comes from human inspection: anomalies are identified by comparing a query image with a defect-free reference. Inspired by this, we propose InspectorGPT, a VLM framework centered on comparative reasoning. Given a normal reference and a query image, InspectorGPT compares them to identify discrepancies and perform multiple inspection tasks with detailed reasoning. We internalize this capability through Chain-of-Thought (CoT) fine-tuning and Group Relative Policy Optimization (GRPO) with tailored, verifiable rewards. We further introduce InspectorGPT-Seg for pixel-level anomaly masks. Segmentation supervision improves anomaly discrimination but weakens semantic reasoning, while joint training fails to balance them. We therefore train the two branches separately and combine them through task-vector fusion. Extensive experiments demonstrate superior multi-dimensional performance and generalization to unseen benchmarks, validating comparative reasoning for comprehensive industrial inspection.

Problem

• industrial anomaly detection
• generalization
• vision-language models
• zero-shot detection
• pixel-level segmentation

Innovation

• comparative reasoning
• Chain-of-Thought fine-tuning
• Group Relative Policy Optimization
• pixel-level anomaly segmentation
• task-vector fusion