Vol. 6 No. 2 (2026): July−December 2026
Articles

Effects of Prompt Example Count on Money-Laundering Transaction Classification from Textualized Financial-Graph Neighborhoods: An Empirical Study

View PDF
  • Ruming Tang,
  • Haobin Pan,
  • Congrui Yang
Ruming Tang
Data Science, Fudan University, Shanghai, China
Haobin Pan
Computer Application Technology, Institute of Computing Technology, Beijing, China
Congrui Yang
Software Engineering, Carnegie Mellon University, Pittsburgh, PA, USA

Published 2026-09-22

Keywords

  • anti-money laundering,
  • graph verbalization,
  • in-context learning,
  • prompt sensitivity

How to Cite

Tang, R., Pan, H., & Yang, C. (2026). Effects of Prompt Example Count on Money-Laundering Transaction Classification from Textualized Financial-Graph Neighborhoods: An Empirical Study. Spectrum of Research, 6(2). Retrieved from https://spectrumofresearch.com/index.php/sr/article/view/64

Copyright (c) 2026 Spectrum of Research

Creative Commons License

This work is licensed under a Creative Commons Attribution 4.0 International License.

Abstract

Large language models can consume financial-graph records after structural attributes are verbalized, yet the effect of demonstration count remains poorly isolated. This study tests whether balanced demonstrations improve money-laundering classification from compact one-hop narratives. We audit IBM AML HI-Small, construct narratives from target transactions and strictly earlier account activity, and compare zero-, one-, three-, and five-shot prompting, where a shot denotes one demonstration per class. A time-held-out sample contains 20 licit and 20 illicit transactions; three demonstration sets assess sensitivity. Qwen2.5-0.5B-Instruct scores the single-token labels legal and illegal through closed-set next-token likelihood. One-shot prompting raises PR-AUC from 0.516 to 0.578, but its paired difference of 0.062 has a 95% bootstrap interval of [-0.105, 0.243]. Three- and five-shot prompting reduce PR-AUC to 0.477 and 0.431 while increasing mean input length from 190 tokens to 916 and 1,395 tokens. Every condition classifies all 40 cases as illegal, yielding 0.500 precision, 1.000 recall, and 0.667 F1. Evidence for a one-shot ranking benefit is limited; additional demonstrations neither correct label bias nor justify their token cost.

View PDF

References

  1. Altman, E., Blanuša, J., von Niederhäusern, L., Egressy, B., Anghel, A. S., & Atasu, K. (2023). Realistic synthetic financial transactions for anti-money laundering models. Advances in Neural Information Processing Systems, 36.
  2. Fatemi, B., Halcrow, J., & Perozzi, B. (2024). Talk like a graph: Encoding graphs for large language models. International Conference on Learning Representations.
  3. Li, Y., Hu, J., Hooi, B., He, B., & Chen, C. (2026). DGP: A dual-granularity prompting framework for fraud detection with graph-enhanced LLMs. Proceedings of the AAAI Conference on Artificial Intelligence, 40, 15171-15179.
  4. Wang, H., Feng, S., He, T., Tan, Z., Han, X., & Tsvetkov, Y. (2023). Can language models solve graph problems in natural language? Advances in Neural Information Processing Systems, 36.
  5. Wei, J., Wang, X., Schuurmans, D., Bosma, M., Ichter, B., Xia, F., Chi, E. H., Le, Q. V., & Zhou, D. (2022). Chain-of-thought prompting elicits reasoning in large language models. Advances in Neural Information Processing Systems, 35.
  6. Wang, X., Wei, J., Schuurmans, D., Le, Q. V., Chi, E. H., Narang, S., Chowdhery, A., & Zhou, D. (2023). Self-consistency improves chain of thought reasoning in language models. International Conference on Learning Representations.
  7. Zhou, Z., Tao, R., Zhu, J., Luo, Y., Wang, Z., & Han, B. (2024). Can language models perform robust reasoning in chain-of-thought prompting with noisy rationales? Advances in Neural Information Processing Systems, 37.
  8. Tang, J., Yang, Y., Wei, W., Shi, L., Su, L., Cheng, S., Yin, D., & Huang, C. (2024). GraphGPT: Graph instruction tuning for large language models. In Proceedings of the 47th International ACM SIGIR Conference on Research and Development in Information Retrieval (pp. 491-500). https://doi.org/10.1145/3626772.3657775
  9. Liu, Y., Ao, X., Qin, Z., Chi, J., Feng, J., Yang, H., & He, Q. (2021). Pick and choose: A GNN-based imbalanced learning approach for fraud detection. In Proceedings of the Web Conference 2021 (pp. 3168-3177). https://doi.org/10.1145/3442381.3449989
  10. Xiang, S., Zhu, M., Cheng, D., Li, E., Zhao, R., Ouyang, Y., Chen, L., & Zheng, Y. (2023). Semi-supervised credit card fraud detection via attribute-driven graph representation. Proceedings of the AAAI Conference on Artificial Intelligence, 37(12), 14557-14565. https://doi.org/10.1609/aaai.v37i12.26702
  11. Huang, X., Yang, Y., Wang, Y., Wang, C., Zhang, Z., Xu, J., Chen, L., & Vazirgiannis, M. (2022). DGraph: A large-scale financial dataset for graph anomaly detection. Advances in Neural Information Processing Systems, 35, 22765-22777.
  12. Elmougy, Y., & Liu, L. (2023). Demystifying fraudulent transactions and illicit nodes in the Bitcoin network for financial forensics. In Proceedings of the 29th ACM SIGKDD Conference on Knowledge Discovery and Data Mining (pp. 3979-3990). https://doi.org/10.1145/3580305.3599803
  13. Dou, Y., Liu, Z., Sun, L., Deng, Y., Peng, H., & Yu, P. S. (2020). Enhancing graph neural network-based fraud detectors against camouflaged fraudsters. In Proceedings of the 29th ACM International Conference on Information & Knowledge Management (pp. 315-324). https://doi.org/10.1145/3340531.3411903
  14. Liu, Z., Dou, Y., Deng, Y., Peng, H., & Yu, P. S. (2020). Alleviating the inconsistency problem of applying graph neural network to fraud detection. In Proceedings of the 43rd International ACM SIGIR Conference on Research and Development in Information Retrieval (pp. 1569-1572). https://doi.org/10.1145/3397271.3401253
  15. Xu, F., Wang, N., Wu, H., Wen, X., Zhao, X., & Wan, H. (2024). Revisiting graph-based fraud detection in sight of heterophily and spectrum. Proceedings of the AAAI Conference on Artificial Intelligence, 38(8), 9214-9222. https://doi.org/10.1609/aaai.v38i8.28773
  16. Liu, Y., Li, S., Zheng, Y., Chen, Q., Zhang, C., & Pan, S. (2024). ARC: A generalist graph anomaly detector with in-context learning. Advances in Neural Information Processing Systems, 37.
  17. Li, Y., Zhang, X., & Hao, C. (2026). Graph Neural Network-Based Cross-Market Risk Contagion Analysis between US Equity Sectors and Treasury Yields during the 2023 Regional Banking Stress. 2026 2nd International Conference on Digital Intelligence and Computing
  18. Graph Neural Network-Based Cross-Market Risk Contagion Analysis between U.S. Equity Sectors and Treasury Yields during the 2023 Regional Banking Stress
  19. Tian, Y. (2026). Gradient Boosting-Based Demand Variability Estimation for Improved Safety Stock Calculation in Multi-Echelon Aerospace Spare Parts Inventory. Journal of Science, Innovation & Social Impact, 2(2), 175-186.
  20. Li, Y. & Fu, X. (2026). Comparative Evaluation of Graph Neural Networks for Cross-Market Risk Contagion Path Identification in Multi-Layer Financial Networks. 2026 2nd International Conference on Artificial Intelligence and Advanced
  21. Wu, C., & Pan, Z. (2024). An Integrated Graph Neural Network and Reinforcement Learning Framework for Intelligent Drug Discovery. Journal of Advanced Computing Systems, 4(6), 19-29.
  22. Chen, Y. (2026). Comparing prompting strategies and backbone choices for LLM agents on tabular credit-card fraud detection: An empirical study. Artificial Intelligence and Machine Learning Review, 7(3), 1-11
  23. Li, Y., Zhao, F., & Hu, J. (2026). Identifying Cross-Market Risk Contagion Amplifiers via Graph Attention Networks: Empirical Evidence from US Financial Stress Periods. Journal of Computing Innovations and Applications 4 (1), 164-175
  24. Chen, Y. (2026). Comparing Prompting Strategies and Backbone Choices for LLM Agents on Tabular Credit-Card Fraud Detection: An Empirical Study. Journal of Innovative Technologies 2 (1)
  25. Li, Y., Zhang, X., & Hao, C. (2026). Graph Neural Network-Based Cross-Market Risk Contagion Analysis between US Equity Sectors and Treasury Yields during the 2023 Regional Banking Stress. Journal of Sustainability, Policy, and Practice, 2(4), 154-168.
  26. Huo, H., Li, Y., Wu, C., Wu, X., Tian, Z., & Liu, Y. (2019, September). Semantic segmentation and scene reconstruction for traffic simulation using CNN. In 2019 2nd China Symposium on Cognitive Computing and Hybrid Intelligence (CCHI) (pp. 117-122). IEEE.
  27. Wei, C., Ge, L. & Brooks, N. (2024). Graph-based representation learning for financial fraud and anomaly transaction detection. Journal of Computing Innovations and Applications 2 (1), 153-164
  28. Z. Feng, D. Zhang, Y. Wang (2024). Intraday Liquidity Patterns and Their Implications for Market Risk Assessment: Evidence from Global Equity Markets. Artificial Intelligence and Machine Learning Review
  29. Kang, A., Li, Z., & Meng, S. (2023). AI-Enhanced Risk Identification and Intelligence Sharing Framework for Anti-money laundering in Cross-Border Income Swap Transactions. Journal of Advanced Computing Systems 3 (5), 34-47
  30. Sun, Y. & Tang, R. (2026). A Comparative Evaluation of Post-hoc Explainability Methods for Graph Neural Networks in Node Classification. Journal of Innovative Technologies 2 (2)
  31. Jiang, Z., & Wang, M. (2025). Evaluation and Analysis of Chart Reasoning Accuracy in Multimodal Large Language Models: An Empirical Study on Influencing Factors. Pinnacle Academic Press Proceedings Series, 3, 43-58.
  32. Kang, A., & Ma, X. (2025, July). AI-Based Pattern Recognition and Characteristic Analysis of Cross-Border money laundering Behaviors in Digital Currency Transactions. In 2025 International Conference on Digital Intelligence and Computing Technologies (pp. 1-19). Pinnacle Academic Press.
  33. Chen, Y. & Hu, J. (2026). Graph Neural Network-Based Cascading Disruption Path Identification in Multi-Tier Rare Earth Processing Networks. Journal of Global Engineering Review 4 (1), 99-112
  34. Wu, Z., Feng, Z., & Dong, B. (2024). Optimal feature selection for market risk assessment: A dimensional reduction approach in quantitative finance. Journal of Computing Innovations and Applications 2 (1), 20-31
  35. Kang, A., Xin, J., & Ma, X. (2024). Anomalous cross-border capital flow patterns and their implications for national economic security: An empirical analysis. Journal of Advanced Computing Systems, 4(5), 42-54.
  36. Yiyi, C. (2025). Federated learning-based framework for privacy-protected cross-border financial risk evaluation: Analyzing US-asia investment flows. Journal of Sustainability, Policy, and Practice, 1(4), 50-65
  37. Wu, X., Li, Y., Liu, Y., Pang, S., Wang, L., Wu, C., & Huo, H. (2019, June). Jointly detecting and retrieving vehicles from road image sequences based on CNN. In 2019 IEEE Intelligent Vehicles Symposium (IV) (pp. 2530-2535). IEEE.
  38. Wei, C. & Wu, C. (2024). Credit risk transmission mechanism and prevention strategies in supply chain finance: A core enterprise perspective. Artificial Intelligence and Machine Learning Review 5 (2), 101-115
  39. Chen, Y. (2024). Explainable Attack Path Reasoning for Industrial Control Network Security Based on Knowledge Graphs. Journal of Computing Innovations and Applications 2 (1), 128-139
  40. Rao, GL., Lu, T., Yan, L., & Liu, Y. (2024). A Hybrid LSTM-KNN Framework for Detecting Market Microstructure Anomalies: Evidence from High-Frequency Jump Behaviors in Credit Default Swap Markets. Journal of Knowledge Learning and Science Technology
  41. Yu, M., & Li, Z. (2025). An Empirical Comparison of Discrete Video Tokenization Schemes for Video Question Answering and Video Captioning. Artificial Intelligence and Machine Learning Review, 6(2), 27-50.
  42. Rao, GL., Wang, Z., & Liang, J. (2025). Reinforcement Learning for Pattern Recognition in Cross-Border Financial Transaction Anomalies: A Behavioral Economics Approach to AML. Applied and Computational Engineering 142, 116-127
  43. Zhang, D. & Wang, Y. (2025). AI-driven quality assessment and investment risk identification for carbon credit projects in developing countries. 2025 International Forum on Computing and Intelligent Systems, 76-92
  44. Zhang, D. & Ma, X. (2025). Machine Learning-Based Credit Risk Assessment for Green Bonds: Climate Factor Integration and Default Prediction Analysis. Journal of Sustainability, Policy, and Practice 1 (2), 121-135
  45. Li, Y. & Zhang, S. (2025). Machine Learning-Based Credit Risk Early Warning System for Small and Medium-Sized Financial Institutions: An Ensemble Learning Approach with Interpretable Risk Indicators. Journal of Science, Innovation & Social Impact 1 (1), 372-383
  46. Kang, A. & Ma, X. (2025). AI-Based Pattern Recognition and Characteristic Analysis of Cross-Border Money laundering Behaviors in Digital Currency Transactions. Pinnacle Academic Press Proceedings Series 5, 1-19
  47. Wu, C., Guan, H., & Weng, H. (2024). Forecasting Hospital Resource Demand Using Gradient Boosting: An Operational Analytics Approach for Bed Allocation and Patient Flow Management. Journal of Computing Innovations and Applications, 2(1), 74-85.
  48. Cao, H. & Yu, M. (2026). Do Neighborhood-Aggregated Features Improve Gradient Boosting for Node-Level Fraud Detection? An Empirical Study on Two Multi-Relational Graphs. Journal of Selected Topics in Academic Research 2 (1)
  49. Chen, Y. & Tang, T. (2026). Evaluating Prompt Engineering Strategies for Few-Shot Cyber Threat Intelligence Entity and Relation Extraction from Multi-Source Reports. Journal of Science, Innovation & Social Impact 2 (2), 153-164
  50. Trinh, TK. & Zhang, D. (2024). Algorithmic fairness in financial decision-making: Detection and mitigation of bias in credit scoring applications. Journal of Advanced Computing Systems 4 (2), 36-49
  51. Zhao, F., & Tang, T. (2026). AI-Based Sentiment Analysis for Stock Market Prediction: A Systematic Literature Review. Journal of Sustainability, Policy, and Practice, 2(3), 115-124.
  52. Z. Feng, C. Ni, S. Zhou (2025). Option-Implied Information for Forward-Looking Market Risk Assessment: Evidence from Commodity Derivatives Markets. Spectrum of Research
  53. Peng, Y. and Gao, S., 2026. An Empirical Comparison of XGBoost and LightGBM for Capital Expenditure Deviation Prediction in US FERC-Regulated Rate-Base Transmission Projects. Journal of Sustainability, Policy, and Practice, 2(4), pp.90-102.
  54. Rao, GL., Trinh, TK., Chen, Y., Shu, M., & Zheng, S. (2024). Jump Prediction in Systemically Important Financial Institutions' CDS Prices. Spectrum of Research 4 (2)
  55. Lai, J., & Li, Z. (2026). A Comparative Empirical Evaluation of Single-Agent and Multi-Agent LLM Prompting Strategies for Automated Formative Feedback in Education. Journal of Intelligence and Engineering Technology, 1(2), 29-38.
  56. Li, Y. & Ling, Z. (2026). Real-Time Multi-Risk Early Warning for Community Banks: An Application of Ensemble Anomaly Detection and Explainable Artificial Intelligence. Journal of Advanced Computing Systems 6 (2), 15-27
  57. Wang, H. & Sun, Y. (2026). An Empirical Analysis of the Temperature Hyperparameter in Knowledge Distillation for Image Classification Student Networks. Journal of Innovative Technologies 2 (1)
  58. Ge, L. & Rao, G. (2025). MultiStream-FinBERT: A Hybrid Deep Learning Framework for Corporate Financial Distress Prediction Integrating Accounting Metrics, Market Signals, and Textual Disclosures. Pinnacle Academic Press Proceedings Series 3, 107-122
  59. Yuan, D. & Zhang, D. (2025). APAC-sensitive anomaly detection: Culturally-aware AI models for enhanced AML in US securities trading. 2025 International Conference on Computer, AI, Systems and Automation, 108-121
  60. Rao, G., Zheng, S., & Guo, L. (2025). Dynamic Reinforcement Learning for Suspicious Fund Flow Detection: A Multi-layer Transaction Network Approach with Adaptive Strategy Optimization. Applied and Computational Engineering 145, 1-11
  61. Li, Y. & Mo, Z. (2026). Backtesting Value-at-Risk and Expected Shortfall Across Calm and Turbulent Markets: An Empirical Comparison of Historical Simulation, GARCH-Family, Filtered Historical Simulation. Annals of Applied Sciences 7 (1)
  62. Li, Y. (2026). Enhancing Financial Compliance Transparency through Automated Data Governance and Intelligent Risk Reporting. Journal of Science, Innovation & Social Impact 2 (1), 299-313
  63. Y. Li, S. Fan, H. Wang (2025). Machine Learning-Based Identification of Anomalous Trading Behavior Patterns Among Asia-Pacific Investors in U.S. Securities Markets. Spectrum of Research
  64. Ming, Y., & Fu, H. (2020, July). Variational bayesian sparsification for distillation compression. In 2020 IEEE International Conference on Multimedia and Expo (ICME) (pp. 1-6). IEEE.
  65. Shang, W., Zhao, F., & Liu, M. (2026). A Comparative Study of Spatiotemporal Clustering and Classification Approaches for Security Incident Risk Assessment in UN Peacekeeping Operations. Journal of Sustainability, Policy, and Practice, 2(3), 1-11.
  66. Liu, W., Rao, G., & Lian, H. (2023). Anomaly Pattern Recognition and Risk Control in High-Frequency Trading Using Reinforcement Learning. Journal of Computing Innovations and Applications 1 (2), 47-58
  67. Wu, Z., Zhang, Z., Zhao, Q., & Yan, L. (2025). Privacy-preserving financial transaction pattern recognition: A differential privacy approach. Preprints
  68. Zhang, R., Fan, C., Ming, Y., Fu, H., & Meng, X. (2020, October). An Effective Hierarchical Resolution Learning Method for Low-Resolution Targets Tracking. In 2020 IEEE International Conference on Image Processing (ICIP) (pp. 2076-2080). IEEE.
  69. Li, Y. & Fu, H. (2026). Interpretable Gradient Boosting for Transmission Capital Expenditure Deviation: A Multi-Decade Benchmark on FERC-Regulated Utilities. Academia Nexus Journal 5 (2)
  70. Sun, M., & Wang, H. (2025). AI-powered customer engagement sequence analysis and conversion funnel optimization in multi-channel e-commerce marketing. Journal of Industrial Engineering and Applied Science, 3(5), 7-20
  71. Ge, L., & Rao, G. (2025, July). MultiStream-FinBERT: A Hybrid Deep Learning Framework for Corporate Financial Distress Prediction Integrating Accounting Metrics, Market Signals, and Textual Disclosures. In 2025 International Forum on Computing and Intelligent Systems (pp. 107-122). Pinnacle Academic Press.
  72. Cao, H. (2024). Privacy-Preserving Click Pattern Anomaly Detection for Mobile In-App Browser Advertising Fraud. COMPUTING INNOVATIONS AND APPLICATIONS 2 (2), 151-161
  73. Wu, Z., Wang, S., Ni, C., & Wu, J. (2024). Adaptive traffic signal timing optimization using deep reinforcement learning in urban networks. Artificial Intelligence and Machine Learning Review 5 (4), 55-68
  74. Cao, H., & Liu, M. (2026). Empirical Evaluation of Constraint Discovery Techniques for Business Logic Consistency Verification in Payroll Data Migration. Journal of Sustainability, Policy, and Practice, 2(3), 92-103.
  75. Does an LLM-Based Feedback Agent Move the Needle? An Empirical Study of Student Correctness and Hint Dependency on the ASSISTments 2017 Interaction Logs
  76. Kang, A., Zhang, K., & Chen, Y. (2025). AI-Assisted Analysis of Policy Communication during Economic Crises: Correlations with Market Confidence and Recovery Outcomes. Pinnacle Academic Press Proceedings Series 3, 159-173
  77. Tang, T., & Yu, M. (2024). A Comparative Evaluation of LLM-Generated Semantic Tags versus Classical Text Features (TF-IDF, LDA, BERT Embeddings) for User-Interest Enrichment in Short-Video Recommendation. Artificial Intelligence and Machine Learning Review, 5(1), 129-140.
  78. Min, S. & Wei, C. (2023). Comparative analysis of filter-based feature selection methods for high-dimensional data in classification tasks. Journal of Advanced Computing Systems 3 (8), 25-38