LAD-Reasoner: Tiny Multimodal Models are Good Reasoners for Logical Anomaly Detection

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Li, Weijia, Chu, Guanglei, Chen, Jiong, Xie, Guo-Sen, Shan, Caifeng, Zhao, Fang
Natura: Preprint
Pubblicazione: 2025
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866913797382340608
author Li, Weijia
Chu, Guanglei
Chen, Jiong
Xie, Guo-Sen
Shan, Caifeng
Zhao, Fang
author_facet Li, Weijia
Chu, Guanglei
Chen, Jiong
Xie, Guo-Sen
Shan, Caifeng
Zhao, Fang
contents Recent advances in industrial anomaly detection have highlighted the need for deeper logical anomaly analysis, where unexpected relationships among objects, counts, and spatial configurations must be identified and explained. Existing approaches often rely on large-scale external reasoning modules or elaborate pipeline designs, hindering practical deployment and interpretability. To address these limitations, we introduce a new task, Reasoning Logical Anomaly Detection (RLAD), which extends traditional anomaly detection by incorporating logical reasoning. We propose a new framework, LAD-Reasoner, a customized tiny multimodal language model built on Qwen2.5-VL 3B. Our approach leverages a two-stage training paradigm that first employs Supervised Fine-Tuning (SFT) for fine-grained visual understanding, followed by Group Relative Policy Optimization (GRPO) to refine logical anomaly detection and enforce coherent, human-readable reasoning. Crucially, reward signals are derived from both the detection accuracy and the structural quality of the outputs, obviating the need for building chain of thought (CoT) reasoning data. Experiments on the MVTec LOCO AD dataset show that LAD-Reasoner, though significantly smaller, matches the performance of Qwen2.5-VL-72B in accuracy and F1 score, and further excels in producing concise and interpretable rationales. This unified design reduces reliance on large models and complex pipelines, while offering transparent and interpretable insights into logical anomaly detection. Code and data will be released.
format Preprint
id arxiv_https___arxiv_org_abs_2504_12749
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle LAD-Reasoner: Tiny Multimodal Models are Good Reasoners for Logical Anomaly Detection
Li, Weijia
Chu, Guanglei
Chen, Jiong
Xie, Guo-Sen
Shan, Caifeng
Zhao, Fang
Computer Vision and Pattern Recognition
Recent advances in industrial anomaly detection have highlighted the need for deeper logical anomaly analysis, where unexpected relationships among objects, counts, and spatial configurations must be identified and explained. Existing approaches often rely on large-scale external reasoning modules or elaborate pipeline designs, hindering practical deployment and interpretability. To address these limitations, we introduce a new task, Reasoning Logical Anomaly Detection (RLAD), which extends traditional anomaly detection by incorporating logical reasoning. We propose a new framework, LAD-Reasoner, a customized tiny multimodal language model built on Qwen2.5-VL 3B. Our approach leverages a two-stage training paradigm that first employs Supervised Fine-Tuning (SFT) for fine-grained visual understanding, followed by Group Relative Policy Optimization (GRPO) to refine logical anomaly detection and enforce coherent, human-readable reasoning. Crucially, reward signals are derived from both the detection accuracy and the structural quality of the outputs, obviating the need for building chain of thought (CoT) reasoning data. Experiments on the MVTec LOCO AD dataset show that LAD-Reasoner, though significantly smaller, matches the performance of Qwen2.5-VL-72B in accuracy and F1 score, and further excels in producing concise and interpretable rationales. This unified design reduces reliance on large models and complex pipelines, while offering transparent and interpretable insights into logical anomaly detection. Code and data will be released.
title LAD-Reasoner: Tiny Multimodal Models are Good Reasoners for Logical Anomaly Detection
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2504.12749