FraudTransformer: Time-Aware GPT for Transaction Fraud Detection

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Aminian, Gholamali, Elliott, Andrew, Li, Tiger, Wong, Timothy Cheuk Hin, Dehon, Victor Claude, Szpruch, Lukasz, Maple, Carsten, Read, Christopher, Brown, Martin, Reinert, Gesine, Mamouei, Mo
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866911235949199360
author Aminian, Gholamali
Elliott, Andrew
Li, Tiger
Wong, Timothy Cheuk Hin
Dehon, Victor Claude
Szpruch, Lukasz
Maple, Carsten
Read, Christopher
Brown, Martin
Reinert, Gesine
Mamouei, Mo
author_facet Aminian, Gholamali
Elliott, Andrew
Li, Tiger
Wong, Timothy Cheuk Hin
Dehon, Victor Claude
Szpruch, Lukasz
Maple, Carsten
Read, Christopher
Brown, Martin
Reinert, Gesine
Mamouei, Mo
contents Detecting payment fraud in real-world banking streams requires models that can exploit both the order of events and the irregular time gaps between them. We introduce FraudTransformer, a sequence model that augments a vanilla GPT-style architecture with (i) a dedicated time encoder that embeds either absolute timestamps or inter-event values, and (ii) a learned positional encoder that preserves relative order. Experiments on a large industrial dataset -- tens of millions of transactions and auxiliary events -- show that FraudTransformer surpasses four strong classical baselines (Logistic Regression, XGBoost and LightGBM) as well as transformer ablations that omit either the time or positional component. On the held-out test set it delivers the highest AUROC and PRAUC.
format Preprint
id arxiv_https___arxiv_org_abs_2509_23712
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle FraudTransformer: Time-Aware GPT for Transaction Fraud Detection
Aminian, Gholamali
Elliott, Andrew
Li, Tiger
Wong, Timothy Cheuk Hin
Dehon, Victor Claude
Szpruch, Lukasz
Maple, Carsten
Read, Christopher
Brown, Martin
Reinert, Gesine
Mamouei, Mo
Machine Learning
Detecting payment fraud in real-world banking streams requires models that can exploit both the order of events and the irregular time gaps between them. We introduce FraudTransformer, a sequence model that augments a vanilla GPT-style architecture with (i) a dedicated time encoder that embeds either absolute timestamps or inter-event values, and (ii) a learned positional encoder that preserves relative order. Experiments on a large industrial dataset -- tens of millions of transactions and auxiliary events -- show that FraudTransformer surpasses four strong classical baselines (Logistic Regression, XGBoost and LightGBM) as well as transformer ablations that omit either the time or positional component. On the held-out test set it delivers the highest AUROC and PRAUC.
title FraudTransformer: Time-Aware GPT for Transaction Fraud Detection
topic Machine Learning
url https://arxiv.org/abs/2509.23712