Multi-Scale Temporal Graph Transformer for Blockchain Anti-Money Laundering
Keywords:
anti-money laundering, blockchain, graph transformer, temporal graph learning, multi-scale modeling, subgraph classificationAbstract
Money laundering on public blockchains is a temporal pattern: automated peeling moves value within minutes, coordinated deposits arrive within the same quarter of an hour, and layering drips funds through intermediaries over days, while licit activity produces the same shapes at other speeds. We present MS-TGT, a multi-scale temporal graph transformer for subgraph-level anti-money laundering that builds on the time-biased subgraph transformer STEMP. MS-TGT describes every account by a multi-scale flow spectrum, i.e., Laplace-domain samples of its residence-time, synchrony and dormancy distributions at six scales from five minutes to four weeks; it biases attention with a signed mixture of exponential time kernels that can express band-pass and causal preferences; and it reads out the subgraph through scale-specific tokens fused by a gate. An analysis of 79,161 real Ethereum phishing transfers shows residence times spread from seconds to months. On a benchmark calibrated with these timings, in which every laundering motif has licit look-alikes with identical topology and amounts, MS-TGT reaches an illicit-class F1 of 0.949 against 0.815 for the strongest of ten baselines, with bootstrap and McNemar tests confirming the difference, and it degrades least under slowed or delayed laundering. Ablations attribute most of the gain to the flow spectrum, which also lifts GIN from 0.690 to 0.875. On a real-data probe with randomized reference models the gains are smaller and explicit gap statistics remain competitive.
Downloads
References
[1] S. Nakamoto, “Bitcoin: A peer-to-peer electronic cash system,” 2008. [Online]. Available: https://bitcoin.org/bitcoin.pdf
[2] M. Weber, G. Domeniconi, J. Chen, D. K. I. Weidele, C. Bellei, T. Robinson, and C. E. Leiserson, “Anti-money laundering in Bitcoin: Experimenting with graph convolutional networks for financial forensics,” in Proc. KDD Workshop Anomaly Detection in Finance, 2019, arXiv:1908.02591.
[3] Y. Elmougy and L. Liu, “Demystifying fraudulent transactions and illicit nodes in the Bitcoin network for financial forensics,” in Proc. 29th ACM SIGKDD Conf. Knowl. Discov. Data Mining (KDD), 2023, pp. 3979–3990, doi: 10.1145/3580305.3599803.
[4] C. Bellei, M. Xu, R. Phillips, T. Robinson, M. Weber, T. Kaler, C. E. Leiserson, Arvind, and J. Chen, “The shape of money laundering: Subgraph representation learning on the blockchain with the Elliptic2 dataset,” 2024, arXiv:2404.19109.
[5] S. Meiklejohn, M. Pomarole, G. Jordan, K. Levchenko, D. McCoy, G. M. Voelker, and S. Savage, “A fistful of bitcoins: Characterizing payments among men with no names,” in Proc. Internet Meas. Conf. (IMC), 2013, pp. 127–140, doi: 10.1145/2504730.2504747.
[6] M. Möser, R. Böhme, and D. Breuker, “An inquiry into money laundering tools in the Bitcoin ecosystem,” in Proc. APWG eCrime Researchers Summit, 2013, pp. 1–14, doi: 10.1109/eCRS.2013.6805780.
[7] F. Zhou, “Moment-constrained side-force allocation for wind-assisted merchant ships: Hydrodynamic performance of retractable anti-leeway fins and flap rudders,” Innovation and Technology Studies, vol. 3, no. 1, pp. 27–41, 2026.
[8] T. N. Kipf and M. Welling, “Semi-supervised classification with graph convolutional networks,” in Proc. Int. Conf. Learn. Represent. (ICLR), 2017.
[9] P. Veličković, G. Cucurull, A. Casanova, A. Romero, P. Liò, and Y. Bengio, “Graph attention networks,” in Proc. Int. Conf. Learn. Represent. (ICLR), 2018.
[10] K. Xu, W. Hu, J. Leskovec, and S. Jegelka, “How powerful are graph neural networks?” in Proc. Int. Conf. Learn. Represent. (ICLR), 2019.
[11] J. Chen, X. Zhang, and X. Han, “Explainable identification of tax-related financial reporting anomalies in U.S. public companies: Cross-fitted tax-consistency residuals and Owen-value Shapley propagation,” World Journal of Economics and Business Research, vol. 1, no. 1, pp. 33–45, 2023.
[12] D. Xu, C. Ruan, E. Korpeoglu, S. Kumar, and K. Achan, “Inductive representation learning on temporal graphs,” in Proc. Int. Conf. Learn. Represent. (ICLR), 2020.
[13] E. Rossi, B. Chamberlain, F. Frasca, D. Eynard, F. Monti, and M. Bronstein, “Temporal graph networks for deep learning on dynamic graphs,” 2020, arXiv:2006.10637.
[14] C. Ying, T. Cai, S. Luo, S. Zheng, G. Ke, D. He, Y. Shen, and T.-Y. Liu, “Do transformers really perform badly for graph representation?” in Proc. Adv. Neural Inf. Process. Syst. (NeurIPS), vol. 34, 2021, pp. 28877–28888.
[15] L. Rampášek, M. Galkin, V. P. Dwivedi, A. T. Luu, G. Wolf, and D. Beaini, “Recipe for a general, powerful, scalable graph transformer,” in Proc. Adv. Neural Inf. Process. Syst. (NeurIPS), vol. 35, 2022, pp. 14501–14515.
[16] U. Alon and E. Yahav, “On the bottleneck of graph neural networks and its practical implications,” in Proc. Int. Conf. Learn. Represent. (ICLR), 2021.
[17] Y. Lin, X. Yue, M. Bian, R. Zhu, and D. Teng, “Subgraph-level temporal graph transformer with laundering-motif prototypes for anti-money laundering on blockchains,” IEEE Access, vol. 14, pp. 139135–139159, 2026, doi: 10.1109/ACCESS.2026.3731160.
[18] A. Paranjape, A. R. Benson, and J. Leskovec, “Motifs in temporal networks,” in Proc. 10th ACM Int. Conf. Web Search Data Mining (WSDM), 2017, pp. 601–610, doi: 10.1145/3018661.3018731.
[19] L. Gauvin, M. Génois, M. Karsai, M. Kivelä, T. Takaguchi, E. Valdano, and C. L. Vestergaard, “Randomized reference models for temporal networks,” SIAM Rev., vol. 64, no. 4, pp. 763–830, 2022, doi: 10.1137/19M1242252.
[20] E. Altman, J. Blanuša, L. von Niederhäusern, B. Egressy, A. Anghel, and K. Atasu, “Realistic synthetic financial transactions for anti-money laundering models,” in Proc. Adv. Neural Inf. Process. Syst. (NeurIPS), Datasets and Benchmarks Track, vol. 36, 2023.
[21] X. Li, S. Liu, Z. Li, X. Han, C. Shi, B. Hooi, H. Huang, and X. Cheng, “FlowScope: Spotting money laundering based on graphs,” in Proc. AAAI Conf. Artif. Intell., vol. 34, no. 4, 2020, pp. 4731–4738, doi: 10.1609/aaai.v34i04.5906.
[22] A. Pareja, G. Domeniconi, J. Chen, T. Ma, T. Suzumura, H. Kanezashi, T. Kaler, T. B. Schardl, and C. E. Leiserson, “EvolveGCN: Evolving graph convolutional networks for dynamic graphs,” in Proc. AAAI Conf. Artif. Intell., vol. 34, no. 4, 2020, pp. 5363–5370, doi: 10.1609/aaai.v34i04.5984.
[23] J. Chen, X. Han, X. Zhang, and Y. Yang, “Exact concept-level explanations for early detection of financial misstatements,” Eurasia Journal of Science and Technology, vol. 5, no. 2, pp. 34–48, 2023.
[24] P. Holme and J. Saramäki, “Temporal networks,” Phys. Rep., vol. 519, no. 3, pp. 97–125, 2012, doi: 10.1016/j.physrep.2012.03.001.
[25] A.-L. Barabási, “The origin of bursts and heavy tails in human dynamics,” Nature, vol. 435, no. 7039, pp. 207–211, 2005, doi: 10.1038/nature03459.
[26] A. Vaswani, N. Shazeer, N. Parmar, J. Uszkoreit, L. Jones, A. N. Gomez, Ł. Kaiser, and I. Polosukhin, “Attention is all you need,” in Proc. Adv. Neural Inf. Process. Syst. (NeurIPS), vol. 30, 2017, pp. 5998–6008.
[27] S. G. Mallat, “A theory for multiresolution signal decomposition: The wavelet representation,” IEEE Trans. Pattern Anal. Mach. Intell., vol. 11, no. 7, pp. 674–693, Jul. 1989, doi: 10.1109/34.192463.
[28] T.-Y. Lin, P. Goyal, R. Girshick, K. He, and P. Dollár, “Focal loss for dense object detection,” in Proc. IEEE Int. Conf. Comput. Vis. (ICCV), 2017, pp. 2999–3007, doi: 10.1109/ICCV.2017.324.
[29] T. Chen and C. Guestrin, “XGBoost: A scalable tree boosting system,” in Proc. 22nd ACM SIGKDD Int. Conf. Knowl. Discov. Data Mining (KDD), 2016, pp. 785–794, doi: 10.1145/2939672.2939785.
[30] J. Demšar, “Statistical comparisons of classifiers over multiple data sets,” J. Mach. Learn. Res., vol. 7, pp. 1–30, 2006.
[31] X. Yue and R. Zhu, “An explainable hybrid machine learning framework for cash-flow-conditioned multi-horizon liquidity risk early warning: An SME-oriented benchmark study,” World Journal of Management Science, vol. 2, no. 1, pp. 63–72, 2024.
[32] J. Ding et al., “Game-Theoretic Cost-Sensitive Adversarial Training for Robust Cloud Intrusion Detection Against GAN-Based Evasion Attacks,” Applied Sciences, vol. 16, art. 3944, 2026, doi: 10.3390/app16083944.
Downloads
Published
Issue
Section
License
Copyright (c) 2026 Isaac Edwards, Benjamin Murphy, Zain Sharma, Diego Lee
This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.
This article is licensed under the Creative Commons Attribution 4.0 International License. This license permits use, sharing, and adaptation of the article for non-commercial purposes, provided the original authors and source are properly credited.
Alternatively, you may choose other Creative Commons licenses based on your preferences:
-
CC Attribution-NonCommercial-NoDerivatives 4.0: This license allows others to download the works and share them with others as long as they credit the author, but they can’t change them in any way or use them commercially.
-
CC Attribution-NonCommercial 4.0: This license allows others to remix, tweak, and build upon the work non-commercially, as long as they credit the author, but they cannot use the material for commercial purposes.
-
CC Attribution-NonCommercial-ShareAlike 4.0: This allows others to remix, tweak, and build upon the work non-commercially, as long as they credit the author and license their new creations under identical terms.
These terms ensure that your article is freely accessible and usable, while respecting the rights of the original authors.
