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

• URL: http://arxiv.org/abs/2608.29783v1
• Date: Sun, 30 Aug 2026 13:22:42 GMT
• Title: InspectorGPT: A Comparative Reasoning Enhanced VLM for Comprehensive Industrial Anomaly Detection
• Abstract summary: Industrial anomaly detection is a critical component of modern manufacturing.<n>We propose InspectorGPT, a VLM framework centered on comparative reasoning.<n>We demonstrate superior multi-dimensional performance and generalization to unseen benchmarks.
• Score: 11.91902831370527
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• 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.

Related papers

• AnomalyAgent: Training-Free Agentic Models for Zero-/Few-Shot Anomaly Detection [43.187781277950386]
  We propose textbfAnomalyAgent, a training-free agentic framework for anomaly detection.<n>Key ingredients include textbf1) a comprehensive anomaly-centric toolset, and textbf2) a customized memory module.<n>Experiment results demonstrate that our AnomalyAgent achieves substantially better performance compared to training-free VLM-based AD and generic agentic methods.
  arXiv  Detail & Related papers  (2026-05-28T16:05:42Z)
• ZSG-IAD: A Multimodal Framework for Zero-Shot Grounded Industrial Anomaly Detection [14.275030421757867]
  ZSG-IAD is a framework for zero-shot grounded industrial anomaly detection.<n>It generates structured anomaly reports and pixel-level anomaly masks.<n>We will release code and annotations to support future research on trustworthy industrial anomaly detection systems.
  arXiv  Detail & Related papers  (2026-04-20T08:30:09Z)
• AgentIAD: Tool-Augmented Single-Agent for Industrial Anomaly Detection [9.208080780536504]
  AgentIAD is a tool-driven framework that enables multi-stage visual inspection.<n>AgentIAD achieves a new state-of-the-art 97.62% classification accuracy on MMAD.
  arXiv  Detail & Related papers  (2025-12-15T18:57:04Z)
• Labels Matter More Than Models: Quantifying the Benefit of Supervised Time Series Anomaly Detection [56.302586730134806]
  Time series anomaly detection (TSAD) is a critical data mining task often constrained by label scarcity.<n>Current research predominantly focuses on Unsupervised Time-series Anomaly Detection.<n>This paper challenges the premise that architectural complexity is the optimal path for TSAD.
  arXiv  Detail & Related papers  (2025-11-20T08:32:49Z)
• Contamination Detection for VLMs using Multi-Modal Semantic Perturbation [73.76465227729818]
  Open-source Vision-Language Models (VLMs) have achieved state-of-the-art performance on benchmark tasks.<n>Pretraining corpora raise a critical concern for both practitioners and users: inflated performance due to test-set leakage.<n>We show that existing detection approaches either fail outright or exhibit inconsistent behavior.<n>We propose a novel simple yet effective detection method based on multi-modal semantic perturbation.
  arXiv  Detail & Related papers  (2025-11-05T18:59:52Z)
• Advancing Machine-Generated Text Detection from an Easy to Hard Supervision Perspective [108.30620357325559]
  Existing machine-generated text (MGT) detection methods implicitly assume labels as the "golden standard"<n>We propose an easy-to-hard enhancement framework to provide reliable supervision under such inexact conditions.
  arXiv  Detail & Related papers  (2025-11-02T15:59:31Z)
• Seeing the Unseen: Towards Zero-Shot Inspection for Wind Turbine Blades using Knowledge-Augmented Vision Language Models [10.230967860299504]
  We propose a zero-shot-oriented inspection framework that integrates Retrieval-Augmented Generation with Vision-Language Models.<n>A multimodal knowledge base is constructed, comprising technical documentation, representative reference images, and domain-specific guidelines.<n>We evaluate the framework on 30 labeled blade images covering diverse damage categories.
  arXiv  Detail & Related papers  (2025-10-26T23:19:28Z)
• Unified Unsupervised Anomaly Detection via Matching Cost Filtering [113.43366521994396]
  Unsupervised anomaly detection (UAD) aims to identify image- and pixel-level anomalies using only normal training data.<n>We present Unified Cost Filtering (UCF), a generic post-hoc refinement framework for refining anomaly cost volume of any UAD model.
  arXiv  Detail & Related papers  (2025-10-03T03:28:18Z)
• Foundation Visual Encoders Are Secretly Few-Shot Anomaly Detectors [58.75916798814376]
  We develop a few-shot anomaly detector termed FoundAD.<n>We observe that the anomaly amount in an image directly correlates with the difference in the learnt embeddings.<n>The simple operator acts as an effective tool for anomaly detection to characterize and identify out-of-distribution regions in an image.
  arXiv  Detail & Related papers  (2025-10-02T11:53:20Z)
• OmniAD: Detect and Understand Industrial Anomaly via Multimodal Reasoning [76.90511414963265]
  We introduce OmniAD, a framework that unifies anomaly detection and understanding for fine-grained analysis.<n>Visual reasoning provides detailed inspection by leveraging Text-as-Mask.<n>Visual Guided Textual Reasoning conducts comprehensive analysis by integrating visual perception.
  arXiv  Detail & Related papers  (2025-05-28T07:02:15Z)
• Towards Training-free Anomaly Detection with Vision and Language Foundation Models [17.991678161890174]
  Anomaly detection is valuable for real-world applications, such as industrial quality inspection.<n>We introduce LogSAD, a novel multi-modal framework that requires no training for both Logical and Structural Anomaly Detection.
  arXiv  Detail & Related papers  (2025-03-24T04:07:59Z)
• Myriad: Large Multimodal Model by Applying Vision Experts for Industrial Anomaly Detection [86.24898024621008]
  We present a novel large multimodal model applying vision experts for industrial anomaly detection(abbreviated to Myriad)<n>We utilize the anomaly map generated by the vision experts as guidance for LMMs, such that the vision model is guided to pay more attention to anomalous regions.<n>Our proposed method not only performs favorably against state-of-the-art methods, but also inherits the flexibility and instruction-following ability of LMMs in the field of IAD.
  arXiv  Detail & Related papers  (2023-10-29T16:49:45Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.