AI RESEARCH
AG-VAS: Anchor-Guided Zero-Shot Visual Anomaly Segmentation with Large Multimodal Models
arXiv CS.AI
•
ArXi:2603.01305v2 Announce Type: replace-cross Large multimodal models (LMMs) exhibit strong task generalization capabilities, offering new opportunities for zero-shot visual anomaly segmentation (ZSAS). However, existing LMM-based segmentation approaches still face fundamental limitations: anomaly concepts are inherently abstract and context-dependent, lacking stable visual prototypes, and the weak alignment between high-level semantic embeddings and pixel-level spatial features hinders precise anomaly localization.