← Back to all news
arXiv Machine Learning By Yin-Ju Wu, Bi-Cheng Yan, Berlin Chen

MINT: Mitigating Structural Inconsistency via Attentive Message Passing for Financial Anomaly Detection

Read the original on arXiv Machine Learning →

MINT is a graph neural network framework designed to detect illicit accounts in financial transaction networks. It addresses structural inconsistency by using an attentive message‑passing scheme that tracks deviations between nodes in bi‑directional flows and a transaction salience module that regularizes node interactions. To cope with limited labeled data, MINT employs Positive‑Unlabeled learning, leveraging a small set of known illicit nodes and many unlabeled ones, and demonstrates superior performance on three benchmark datasets.

Machine-generated by The Flow from the publisher's headline and feed description — not written or checked by a human. The full article lives at arXiv Machine Learning.

Related stories

arXiv Machine Learning
Oct 22

Leveraging Time-Series Foundation Models for Financial Crime Detection

The paper introduces Temporal-Context Edge Enrichment (TCEE), a framework that uses frozen time‑series foundation models (TSFMs) to summarise each account’s transaction history and enrich transaction representations with temporal context. These enriched edges can be directly scored for classification or used as features in a graph neural network (GNN). Across six AMLWorld datasets, TCEE improves PR‑AUC by an average of 10.8% over the best baseline and trains 6.0× faster; when combined with a GNN, it yields an additional 4.8% improvement on smaller datasets and a 14.5% gain on an Ethereum phishing dataset.

By Danae Natalie Savvidi, H. \c{C}a\u{g}r{\i} Bilgi, Kubilay Atasu, Arie van Deursen
More like this →