EUVIMEDEUVIMED entdecken
Crossref

IAD-R1: Reinforcing Consistent Reasoning in Industrial Anomaly Detection

Yanhui Li, Yunkang Cao, Chengliang Liu, Yuan Xiong, Xinghui Dong, Chao Huang

Proceedings of the AAAI Conference on Artificial Intelligence · 14.03.2026 · 40 (8) · 6583-6591

DOI: 10.1609/aaai.v40i8.37588 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Proceedings of the AAAI Conference on Artificial Intelligence, online erschienen am 14.03.2026. Verfasst von Yanhui Li und 5 weiteren Autor:innen. Die Arbeit wurde bisher 10-mal zitiert (Crossref).

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

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.

Zitieren

APA

Li, Y., Cao, Y., Liu, C., Xiong, Y., Dong, X., & Huang, C. (2026). IAD-R1: Reinforcing Consistent Reasoning in Industrial Anomaly Detection. Proceedings of the AAAI Conference on Artificial Intelligence, 40(8), 6583-6591. https://doi.org/10.1609/aaai.v40i8.37588

Vancouver

Li Y, Cao Y, Liu C, Xiong Y, Dong X, Huang C. IAD-R1: Reinforcing Consistent Reasoning in Industrial Anomaly Detection. Proceedings of the AAAI Conference on Artificial Intelligence. 2026;40(8):6583-6591. doi:10.1609/aaai.v40i8.37588

BibTeX
@article{Li2026,
  author = {Li, Yanhui and Cao, Yunkang and Liu, Chengliang and Xiong, Yuan and Dong, Xinghui and Huang, Chao},
  title = {IAD-R1: Reinforcing Consistent Reasoning in Industrial Anomaly Detection},
  journal = {Proceedings of the AAAI Conference on Artificial Intelligence},
  year = {2026},
  volume = {40},
  number = {8},
  pages = {6583--6591},
  doi = {10.1609/aaai.v40i8.37588}
}
Kostenlos starten →