The paper investigates how many normal samples are required to reliably set an alarm threshold for few‑shot anomaly detectors, focusing on distribution‑free certification limits. Using a frozen DINOv2 PCA residual ranker on 15 MVTec and 12 VisA categories, the authors show that simple leave‑one‑image‑out calibration is limited by resolution and shift, leading to empirical false‑alarm rates far above the nominal level. They derive a category‑count feasibility calculus, demonstrating that at least 14, 29, and 59 independent category draws are needed for 95% upper confidence bounds at α=0.20, 0.10, and 0.05, and propose the CRESS protocol to split source categories into reference, proposal, and certification roles.
whyItMatters":"The study provides concrete numerical thresholds for the amount of source evidence needed to guarantee reliable anomaly detection in new categories, informing practical deployment of few‑shot detectors."
By Gia Huy Thai, Nguyen Thai Anh
arXiv:2608. 15090v1 Announce Type: cross Abstract: Studies of industrial visual inspection commonly report the area under the receiver operating characteristic curve (AUROC) and the overlap between anomaly maps and defect masks.
By Jie Deng
arXiv:2606. 15153v1 Announce Type: new Abstract: Selective prediction with distribution-free risk control promises that, with confidence 1-delta over the calibration draw, the error rate of accepted inputs stays below a user budget alpha.
By Jingwen Zhou, Mingzhe Wang
The paper introduces a framework for auditing image editors that can satisfy regional plausibility constraints while lacking a single latent explanation for the entire output. It defines a common witness grade and witness nerve to formalize local-to-global failures, separating auditing from causal identification. Using Helly-type arguments and a blocker-hypergraph formula, the authors provide short incompatibility certificates, sharp partial-identification bounds for specified image features, and finite-sample outer coverage of identified intervals, validated through controlled MNIST, Morpho-MNIST, and smallNORB experiments.
By Usef Faghihi, Amir Saki
The paper introduces Audit-First VAPO, a method that separates directional admission from magnitude control for selective updates under imperfect verification. It employs an observation-only accept‑appeal‑abstain policy with a finite secondary‑verification budget, freezing the action trace before clean labels are added, and uses finite‑sample bounds to certify harmful risk, coverage, and verifier‑call rate. Experiments on Qwen3.5‑0.8B and GSM8K show that RC‑VAPO achieves high accuracy while maintaining low selected harmful risk and modest verifier cost compared to baseline methods.
By Miaobo Hu, Shuhao Hu, Xiaobo Guo, Xin Wang, Bokun Wang, Rui Chen, Daren Zha
arXiv:2610.00111v1 Announce Type: new
Abstract: Model judges now supervise multimodal systems at scale, filtering training data, selecting outputs, and supplying the reward that shapes multimodal rea...
By Rasul Khanbayov, Hasan Kurban