[2511.08939] TransactionGPT

Computer Science > Machine Learning arXiv:2511.08939 (cs) [Submitted on 12 Nov 2025 (v1), last revised 2 Mar 2026 (this version, v2)] Title:TransactionGPT Authors:Yingtong Dou, Zhimeng Jiang, Tianyi Zhang, Mingzhi Hu, Zhichao Xu, Shubham Jain, Uday Singh Saini, Xiran Fan, Jiarui Sun, Menghai Pan, Junpeng Wang, Xin Dai, Liang Wang, Chin-Chia Michael Yeh, Yujie Fan, Yan Zheng, Vineeth Rakesh, Huiyuan Chen, Guanchu Wang, Mangesh Bendre, Zhongfang Zhuang, Xiaoting Li, Prince Aboagye, Vivian Lai, Minghua Xu, Hao Yang, Yiwei Cai, Mahashweta Das, Yuzhong Chen View a PDF of the paper titled TransactionGPT, by Yingtong Dou and 28 other authors View PDF HTML (experimental) Abstract:We present TransactionGPT (TGPT), a foundation model for consumer transaction data within one of the world's largest payment networks. TGPT is designed to understand and generate transaction trajectories while simultaneously supporting a variety of downstream prediction and classification tasks. We introduce a novel 3D-Transformer architecture specifically tailored for capturing the complex dynamics in payment transaction data. This architecture incorporates design innovations that enhance modality fusion and computational efficiency, while seamlessly enabling joint optimization with downstream objectives. Trained on billion-scale real-world transactions, TGPT significantly improves downstream anomaly transaction detection performance against a competitive production model and exhibits advantages over bas...