Res²CLIP proposes a residual-to-residual alignment framework within CLIP's residual space for few-shot anomaly detection that focuses on relative anomaly deviations across visual and text modalities.
Mrad: Zero-shot anomaly detection with memory-driven retrieval
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
fields
cs.CV 2years
2026 2verdicts
UNVERDICTED 2representative citing papers
LAKE identifies sparse anomaly-sensitive neurons in pre-trained VLMs using minimal normal samples to build compact normality representations and achieve SOTA anomaly detection with neuron-level interpretability.
citing papers explorer
-
Res$^2$CLIP: Few-Shot Generalist Anomaly Detection with Residual-to-Residual Alignment
Res²CLIP proposes a residual-to-residual alignment framework within CLIP's residual space for few-shot anomaly detection that focuses on relative anomaly deviations across visual and text modalities.
-
Latent Anomaly Knowledge Excavation: Unveiling Sparse Sensitive Neurons in Vision-Language Models
LAKE identifies sparse anomaly-sensitive neurons in pre-trained VLMs using minimal normal samples to build compact normality representations and achieve SOTA anomaly detection with neuron-level interpretability.