Account takeover (ATO) fraud is a growing threat to digital banking, requiring effective detection while minimizing friction for legitimate customers. Production systems predominantly rely on tabular models that score sessions in isolation, discarding the relational structure of the underlying interaction network. Although graph-based models exploit relationships among sessions and network entities, they primarily reason over local neighborhoods and therefore capture only part of the problem: fraud risk depends jointly on the local relational structure surrounding a session and the evolving global state of the fraud ecosystem. We present HERMES (HEterogeneous Relational Micro--macro graph transformer Encoder for high-risk Sessions), a dual-scale architecture that jointly models these complementary scales of information. Micro-GT captures local heterogeneous graph structure through structured, relation-aware attention over a temporally safe session neighborhood. Complementing this local representation, Macro-GT models ecosystem-level context using non-anticipative climate tokens that summarize fraud dynamics, platform shifts, and infrastructure reuse, together with adaptive class prototypes that track representative fraud and benign session patterns over time. Evaluated on more than 130 million high-risk transaction sessions from a leading U.S. financial institution, HERMES consistently outperforms production and strong graph-based baselines, achieving a 44.44% relative reduction in customer friction and a 24.66% relative improvement in fraud recall over the production system. Ablation and temporal-stability analyses further demonstrate complementary gains from local relational modeling and global ecosystem context across changing fraud regimes.
Figures & tables
Figure 1: Hermes architecture. Micro-GT encodes the target session’s local heterogeneous neighborhood through structured relational attention, while Macro-GT models non-anticipative ecosystem context together with adaptive fraud and benign reference prototypes.
Figure 2: Micro-GT local graph tokenization with the target session, hop-1 and hop-2 predecessors, and target-adjacent time-conditioned identifier states.
Figure 3: Micro-GT applies four sequential attention patterns over the local token sequence followed by seed-conditioned pooling to obtain hmicro .
Model
AUC
P1 AUC
P2 AUC
FR@50R ↓
R@5FR ↑
XGBoost (production)
80.26
78.05
74.10
9.671
39.24
MLP
80.29 (+0.03)
77.99 ( − 0.08)
75.10 (+1.35)
9.360 ( − 3.21)
39.78 (+1.39)
FT-Transformer
79.16 ( − 1.37)
77.02 ( − 1.31)
66.54 ( − 10.20)
11.213 (+15.94)
35.42 ( − 9.74)
Transformer-Seq
81.95 (+2.10)
80.28 (+2.86)
72.97 ( − 1.52)
9.244 ( − 4.41)
39.60 (+0.91)
GraphSAGE
82.00 (+2.16)
80.05 (+2.57)
76.20 (+2.84)
8.701 ( − 10.03)
41.54 (+5.87)
GraphSAGE + LP (ATLAS)
84.02 (+4.69)
82.45 (+5.64)
78.75 (+6.28)
6.325 ( − 34.59)
46.93 (+19.62)
Table 1: Overall performance and component ablation. All values are in %. FR@50R denotes the flag rate at 50% fraud recall ( ↓ ), and R@5FR denotes fraud recall at a 5% flag rate ( ↑ ). Parentheses report relative change from the production XGBoost baseline.
Figure 4: Temporal stability: ROC-AUC of Hermes and key baselines across four consecutive held-out test months. Hermes consistently outperforms all baselines.
Credit card fraud detection typically relies on tabular features, while repeated attributes can also provide useful relational signals. This paper proposes THGT-FD, a Temporal Heterogeneous Graph Transformer for Fraud Detection. Each transaction is represented using one transaction token and six types of relation tokens and incorporates Time2Vec encoding into the transaction representation. A Transformer learns the interactions among these tokens within each individual transaction and then outputs a fraud probability. Experiments were conducted on 150,000 transactions sampled from the IEEE-CIS Fraud Detection dataset and chronologically partitioned according to TransactionDT. On the test set, THGT-FD achieved an AUC-ROC of 0.8536, an average precision of 0.4164, and a Recall@5% of 0.4708. The class-weighted histogram-based gradient-boosting baseline achieved an AUC-ROC of 0.8722. The results indicate that relation tokens provide useful information for fraud-risk ranking, although the current model does not yet incorporate entity-level historical aggregation.
Qinwen Yan
University of California, Los Angeles Los Angeles, California, USA
Financial fraud in corporate transaction networks has grown more coordinated and harder to detect with rule-based engines and with classical learning models that treat each transaction in isolation. This paper presents GTFD, a graph-transformer fraud detector that fuses structural and temporal evidence from a corporation's payment graph. GTFD encodes the graph with a multi-head graph attention network, encodes ordered transaction sequences with a gated transformer, and combines both views through a cross-modal gating layer. A conformal risk-control head converts the fused representation into threshold-free anomaly scores with finite-sample coverage guarantees, and the network is trained with self-supervised link-mask pretraining plus adversarial augmentation so it remains stable under scarce labels and under adversarial perturbation. On a corporate transaction benchmark enriched with coordinated fraud rings, GTFD reaches an AUROC of 0.990, an F1-score of 96.1% (precision 96.3%, recall 95.9%), and an accuracy of 98.4%. It reduces the false-positive rate by about 29% relative to the strongest baseline while raising coordinated fraud-ring recall from 85.1% to 96.5%. Ablations attribute roughly 2.0 AUROC points to self-supervised pretraining and 1.9 AUROC points to the conformal head, and adversarial stress tests show GTFD retains 89.2% accuracy at perturbation magnitude 0.20 where the next-best model falls to 76.4%.
Financial fraud detection in digital banking requires reasoning over multiple heterogeneous event streams -- transactions, login sessions, risk signals -- that individually appear benign but collectively reveal fraudulent patterns. We propose the Multi-Stream Fraud Transformer (MSFT), a unified architecture that encodes each event stream with independent Transformer encoders and fuses their representations through configurable mechanisms. We conduct a systematic ablation study comparing five fusion strategies: concatenation, gated fusion, time-aware positional encoding, cross-stream attention, and a full combination. On a large-scale dataset (10M users, 1.5% fraud rate) with 85M parameter models, we demonstrate that (1) sequence models significantly outperform gradient-boosted trees operating on aggregated features (0.74 vs. 0.99 AUROC), (2) per-stream encoding is essential -- a single-stream Transformer baseline with matched parameter budget reaches only 0.82 AUROC, an 18-point gap that confirms the multi-stream inductive bias is necessary, (3) time-aware positional encoding achieves the highest discrimination (0.9961 AUROC), (4) gated fusion yields the best precision (0.989) suitable for production deployment, and (5) the risk event stream provides the strongest individual signal contribution. We further validate on proprietary production data from a digital banking platform, showing over 22% relative AUROC improvement over the XGBoost baseline.