# RINSE: Robust Target-Time Normality Estimation for Zero-Shot Graph Anomaly Detection

arXiv: https://arxiv.org/abs/2609.02497 · HTML: https://mlsanity.com/paper/2609.02497/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 10 of its 23 references that are in the map.
- First posted on arXiv 2 Sep 2026.

## Builds on (most cited first, 10 of 10)

- [Rethinking Graph Neural Networks for Anomaly Detection](https://mlsanity.com/paper/2205.15508/) (2022)
- [ZeroG: Investigating Cross-dataset Zero-shot Transferability in Graphs](https://mlsanity.com/paper/2402.11235/) (2024)
- [Empowering Graph Representation Learning with Test-Time Graph Transformation](https://mlsanity.com/paper/2210.03561/) (2022)
- [ARC: A Generalist Graph Anomaly Detector with In-Context Learning](https://mlsanity.com/paper/2405.16771/) (2024)
- [Zero-shot Generalist Graph Anomaly Detection with Unified Neighborhood Prompts](https://mlsanity.com/paper/2410.14886/) (2024)
- [A Survey of Generalization of Graph Anomaly Detection: From Transfer Learning to Foundation Models](https://mlsanity.com/paper/2509.06609/) (2025)
- [Correcting False Alarms from Unseen: Adapting Graph Anomaly Detectors at Test Time](https://mlsanity.com/paper/2511.07023/) (2025)
- [E-BATS: Efficient Backpropagation-Free Test-Time Adaptation for Speech Foundation Models](https://mlsanity.com/paper/2506.07078/) (2025)
- [Tractable and Near-Optimal Adversarial Algorithms for Robust Estimation in Contaminated Gaussian Models](https://mlsanity.com/paper/2112.12919/) (2021)
- [OWLEYE: Zero-Shot Learner for Cross-Domain Graph Data Anomaly Detection](https://mlsanity.com/paper/2601.19102/) (2026)

## Abstract (arXiv)

Zero-shot graph anomaly detection seeks to deploy a detector trained on source graphs to unseen, unlabeled targets, yet domain shift can make source-derived notions of normality unreliable. We introduce RINSE (Robust Iterative Normality Self-Estimation), a gradient-free target-time framework that keeps the source-trained detector fixed while sequentially estimating target normality, representation calibration, and evidence reliability from the target graph. Its core idea is to identify a reliable subset of low-residual target nodes, use them to construct a trimmed target-aware normality model, and combine complementary anomaly evidence through reliability-gated rank fusion and encoder ensembling. Across eight unseen target graphs, RINSE achieves the highest average AUPRC among the evaluated methods under two separate preprocessing protocols, while block ablations and sensitivity analyses support the combined design. These results support robust target-time estimation as a practical approach to generalist graph anomaly detection without target labels, gradients, or per-target tuning.