LogicAD detects logical anomalies in industrial images by extracting text features with vision-language models and verifying consistency with a theorem prover, reporting 86.0% AUROC on MVTec LOCO AD in a one-shot setting.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CV 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
LogicAD: Explainable Anomaly Detection via VLM-based Text Feature Extraction
LogicAD detects logical anomalies in industrial images by extracting text features with vision-language models and verifying consistency with a theorem prover, reporting 86.0% AUROC on MVTec LOCO AD in a one-shot setting.