A 300K quadruplet dataset and UniDG foundation model enable reference- or text-driven defect generation across categories, outperforming few-shot baselines on anomaly detection tasks.
Beyond MLLM-based approaches, these contextual signals may also benefit vision-language alignment-based anomaly detection methods (Gao et al., 2026)
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CV 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Large-Scale Universal Defect Generation: Foundation Models and Datasets
A 300K quadruplet dataset and UniDG foundation model enable reference- or text-driven defect generation across categories, outperforming few-shot baselines on anomaly detection tasks.