English(EN) NeighborDiv: Training-free Zero-shot Generalist Graph Anomaly Detection via Neighbor Diversity

新方法应对图和多模态数据中的通用异常检测

作者 PulseAugur 编辑部 · [2 个来源] · 2026-05-15 16:49

两篇新研究论文介绍了通用异常检测的新方法。NeighborDiv 专注于图数据，提出了一种无需训练的方法，该方法分析节点邻居内的多样性，而不是节点与邻居的一致性，并取得了最先进的成果。Res$^2$CLIP 通过在残差空间中对齐多模态表示来解决少样本通用异常检测问题，旨在提高在无需重新训练的情况下对新类别的泛化能力。 AI

影响引入了新的异常检测技术，有望提高各种应用的性能和泛化能力。

排序理由两篇学术论文介绍了新的异常检测方法。

在 arXiv cs.LG 阅读 →

AI 生成摘要 · Google Gemini · 来自 2 个来源。我们如何撰写摘要 →

报道来源 [2]

arXiv cs.LG TIER_1 English(EN) · Yuke Li · 2026-05-20 08:16

NeighborDiv: Training-free Zero-shot Generalist Graph Anomaly Detection via Neighbor Diversity

Graph Anomaly Detection (GAD) is increasingly shifting to Generalist GAD (GGAD) for cross-domain "one-for-all" detection, but existing GGAD methods predominantly rely on the neighbor consistency principle, falling into the \textbf{Node-to-Neighbor Consistency Paradigm} for anomal…
arXiv cs.CV TIER_1 English(EN) · Shuo Zhang · 2026-05-15 16:49

Res$^2$CLIP: Few-Shot Generalist Anomaly Detection with Residual-to-Residual Alignment

Few-shot Generalist Anomaly Detection requires models to generalize to novel categories without retraining, posing significant challenges in real-world scenarios with scarce samples and rapidly changing categories. Existing CLIP-based methods face two major challenges: coarse-gra…

报道来源 [2]

NeighborDiv: Training-free Zero-shot Generalist Graph Anomaly Detection via Neighbor Diversity

Res$^2$CLIP: Few-Shot Generalist Anomaly Detection with Residual-to-Residual Alignment

相关实体

相关话题