JUDO: A Juxtaposed Domain-Oriented Multimodal Reasoner for Industrial Anomaly QA

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Kang, Hyunju, Lee, Woohyun, Kim, Jaewon, Park, Hogun
Natura: Preprint
Pubblicazione: 2026
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866913146671726592
author Kang, Hyunju
Lee, Woohyun
Kim, Jaewon
Park, Hogun
author_facet Kang, Hyunju
Lee, Woohyun
Kim, Jaewon
Park, Hogun
contents Industrial anomaly detection has been significantly advanced by Large Multimodal Models (LMMs), enabling diverse human instructions beyond detection, particularly through visually grounded reasoning for better image understanding. However, LMMs lack domain-specific knowledge, which limits their ability to generate accurate responses in complex industrial scenarios. In this work, we present JUDO, Juxtaposed Domain-Oriented Multimodal Reasoner, a framework that efficiently incorporates domain knowledge and context in visual and textual reasoning. Through visual reasoning, our model segments the defect region by juxtaposing query images with normal images as visual domain context, enabling a fine-grained visual comparative inspection. Furthermore, we inject domain knowledge through supervised fine-tuning (SFT) to enhance context understanding and subsequently guide domain reasoning through reinforcement learning (GRPO) with tailored rewards, opting for a domain-oriented reasoning process. Experimental results demonstrate that JUDO achieves superior performance on the MMAD benchmark, surpassing models such as Qwen2.5-VL-7B and GPT-4o. These results highlight the importance of enhancing domain knowledge and context for effective reasoning in anomaly understanding.
format Preprint
id arxiv_https___arxiv_org_abs_2605_20284
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle JUDO: A Juxtaposed Domain-Oriented Multimodal Reasoner for Industrial Anomaly QA
Kang, Hyunju
Lee, Woohyun
Kim, Jaewon
Park, Hogun
Computer Vision and Pattern Recognition
Artificial Intelligence
Machine Learning
Industrial anomaly detection has been significantly advanced by Large Multimodal Models (LMMs), enabling diverse human instructions beyond detection, particularly through visually grounded reasoning for better image understanding. However, LMMs lack domain-specific knowledge, which limits their ability to generate accurate responses in complex industrial scenarios. In this work, we present JUDO, Juxtaposed Domain-Oriented Multimodal Reasoner, a framework that efficiently incorporates domain knowledge and context in visual and textual reasoning. Through visual reasoning, our model segments the defect region by juxtaposing query images with normal images as visual domain context, enabling a fine-grained visual comparative inspection. Furthermore, we inject domain knowledge through supervised fine-tuning (SFT) to enhance context understanding and subsequently guide domain reasoning through reinforcement learning (GRPO) with tailored rewards, opting for a domain-oriented reasoning process. Experimental results demonstrate that JUDO achieves superior performance on the MMAD benchmark, surpassing models such as Qwen2.5-VL-7B and GPT-4o. These results highlight the importance of enhancing domain knowledge and context for effective reasoning in anomaly understanding.
title JUDO: A Juxtaposed Domain-Oriented Multimodal Reasoner for Industrial Anomaly QA
topic Computer Vision and Pattern Recognition
Artificial Intelligence
Machine Learning
url https://arxiv.org/abs/2605.20284