# ATM-GAD: Adaptive Temporal Motif Graph Anomaly Detection for Financial Transaction Networks

arXiv: https://arxiv.org/abs/2508.20829 · HTML: https://mlsanity.com/paper/2508.20829/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 22 of its 35 references that are in the map.
- First posted on arXiv 28 Aug 2025.

## Builds on (most cited first, 22 of 22)

- [Attention Is All You Need](https://mlsanity.com/paper/1706.03762/) (2017)
- [PyTorch: An Imperative Style, High-Performance Deep Learning Library](https://mlsanity.com/paper/1912.01703/) (2019)
- [Semi-Supervised Classification with Graph Convolutional Networks](https://mlsanity.com/paper/1609.02907/) (2016)
- [Graph Attention Networks](https://mlsanity.com/paper/1710.10903/) (2017)
- [Inductive Representation Learning on Large Graphs](https://mlsanity.com/paper/1706.02216/) (2017)
- [XGBoost: A Scalable Tree Boosting System](https://mlsanity.com/paper/1603.02754/) (2016)
- [Generating Long Sequences with Sparse Transformers](https://mlsanity.com/paper/1904.10509/) (2019)
- [Motifs in Temporal Networks](https://mlsanity.com/paper/1612.09259/) (2016)
- [Graph Anomaly Detection via Multi-Scale Contrastive Learning Networks with Augmented View](https://mlsanity.com/paper/2212.00535/) (2022)
- [Provably Powerful Graph Neural Networks for Directed Multigraphs](https://mlsanity.com/paper/2306.11586/) (2023)
- [Demystifying Fraudulent Transactions and Illicit Nodes in the Bitcoin Network for Financial Forensics](https://mlsanity.com/paper/2306.06108/) (2023)
- [SLADE: Detecting Dynamic Anomalies in Edge Streams without Labels via Self-Supervised Learning](https://mlsanity.com/paper/2402.11933/) (2024)
- [Motif Graph Neural Network](https://mlsanity.com/paper/2112.14900/) (2021)
- [ARISE: Graph Anomaly Detection on Attributed Networks via Substructure Awareness](https://mlsanity.com/paper/2211.15255/) (2022)
- [Using Motif Transitions for Temporal Graph Generation](https://mlsanity.com/paper/2306.11190/) (2023)
- [Financial Default Prediction via Motif-preserving Graph Neural Network with Curriculum Learning](https://mlsanity.com/paper/2403.06482/) (2024)
- [Temporal-Amount Snapshot MultiGraph for Ethereum Transaction Tracking](https://mlsanity.com/paper/2102.08013/) (2021)
- [Scalable Motif Counting for Large-scale Temporal Graphs](https://mlsanity.com/paper/2204.09236/) (2022)
- [Temporal Motifs for Financial Networks: A Study on Mercari, JPMC, and Venmo Platforms](https://mlsanity.com/paper/2301.07791/) (2023)
- [Scalable Temporal Motif Densest Subnetwork Discovery](https://mlsanity.com/paper/2406.10608/) (2024)
- [Motif-Consistent Counterfactuals with Adversarial Refinement for Graph-Level Anomaly Detection](https://mlsanity.com/paper/2407.13251/) (2024)
- [Blockchain: A Graph Primer](https://mlsanity.com/paper/1708.08749/) (2017)

## Built on by (most cited first, 1 of 1)

- [LLMTM: Benchmarking and Optimizing LLMs for Temporal Motif Analysis in Dynamic Graphs](https://mlsanity.com/paper/2512.22266/) (2025)

## Abstract (arXiv)

Financial fraud detection is essential to safeguard billions of dollars, yet the intertwined entities and fast-changing transaction behaviors in modern financial systems routinely defeat conventional machine learning models. Recent graph-based detectors make headway by representing transactions as networks, but they still overlook two fraud hallmarks rooted in time: (1) temporal motifs--recurring, telltale subgraphs that reveal suspicious money flows as they unfold--and (2) account-specific intervals of anomalous activity, when fraud surfaces only in short bursts unique to each entity. To exploit both signals, we introduce ATM-GAD, an adaptive graph neural network that leverages temporal motifs for financial anomaly detection. A Temporal Motif Extractor condenses each account's transaction history into the most informative motifs, preserving both topology and temporal patterns. These motifs are then analyzed by dual-attention blocks: IntraA reasons over interactions within a single motif, while InterA aggregates evidence across motifs to expose multi-step fraud schemes. In parallel, a differentiable Adaptive Time-Window Learner tailors the observation window for every node, allowing the model to focus precisely on the most revealing time slices. Experiments on four real-world datasets show that ATM-GAD consistently outperforms seven strong anomaly-detection baselines, uncovering fraud patterns missed by earlier methods.