Large Language Models for Financial Fraud Detection: A Systematic Review of Methodologies, Performance, and Future Directions | IJCT Volume 13 – Issue 1 | IJCT-V13I1P9
Large Language Models for Financial Fraud Detection: A Systematic Review of Methodologies, Performance, and Future Directions | IJCT Volume 13 – Issue 1 | IJCT-V13I1P9
Financial fraud continues to pose significant threats to global economic stability, with reported losses reaching un- precedented levels in recent years. The traditional detection methods for fraud are valuable; however they suffer from changing fraud patterns, extreme class imbalances and increased complexity in the fraud schemes. This systematic review analyzes 33 peer-reviewed studies, primarily published between 2023 and 2025 with foundational studies from 2010-2013 included for tra- ditional method comparison, examining how Large Language Models (LLMs) are transforming financial fraud detection across multiple domains including credit card transactions, fintech applications, trading systems, and insurance claims. The literature can be grouped into six (6) questions based on the six (6) areas of research that address the detection performance, architec- tural innovation, computing requirements, domain coverage, implementation challenges, and future directions for the area of fraud detection. Our analysis reveals that hybrid systems combining LLMs with traditional machine learning consistently out- perform standalone approaches, while identifying significant research gaps in forex market fraud detection despite its massive daily trading volume. This review presents the most significant advances in technical innovation including; Data Serialization Techniques, Multi-Agent Frameworks, and Retrieval-Augmented Generation Systems, which have helped advance the space of LLM based Fraud Detection. Additionally, this review provides practical guidance on how Financial Institutions can implement LLM solutions for fraud detection, as well as Priority Areas for Future Research such as Real-Time Processing Optimization, Cross-Domain Generalization, and Automated Regulatory Compliance. Ultimately, this review is intended to serve as a starting point for Researchers and Practitioners who wish to gain an understanding of and continue advancing the use of LLMs for fraud detection.
Keywords
Large Language Models, Financial Fraud Detection, Machine Learning, Deep Learning, Systematic Review, Credit Card Fraud, Fintech Security, Transformer Networks.
Conclusion
This systematic review of 33 studies provides comprehensive evidence that Large Language Models have advanced finan- cial fraud detection from conceptual exploration to practical deployment. The analysis addresses 6 research questions, re- vealing that LLM-integrated systems achieve superior detec- tion accuracy compared to traditional approaches, with hy- brid architectures demonstrating consistent performance im- provements. Key findings include: (1) the identification of a critical research gap in forex market fraud detection de- spite the market’s massive daily trading volume; (2) signifi- cant reductions in false positive rates through advanced detec- tion paradigms; (3) the emergence of multi-agent systems and RAG frameworks as promising architectural innovations; and
(4) the persistent challenge of meeting real-time processing re- quirements with pure LLM approaches. Based on the findings, it appears that hybrid architectures that combine Large Lan- guage Models (LLMs) with conventional Machine Learning, are currently the most effective approach to provide an optimal trade-off between accuracy, latency, interpretability and cost- effectiveness in fraud-detection systems. The next step for fu- ture research is to focus on the application of such architectures to the Forex Market, to Ultra-High-Frequency Processing Sys- tems, Cross-Domain Generalization, and Automated Regula- tory Compliance. The systematic review has established LLM- based fraud detection as a potentially transformative ability to protect global financial security. Thus, continued research investment into this area and the practical implementation of these architectures will be critical in ensuring the financial well-being of billions of users around the world.
References
[1]Y. Nie, Y. Kong, X. Dong, J. M. Mulvey, H. V. Poor, Q. Wen, and S. Zohren, “A survey of large language models for financial applications: Progress, prospects and chal- lenges,” arXiv preprint arXiv:2406.11903, 2024.
[2]
X. Wang, H. Dai, Z. Huang, and Y. Han, “Network traffic anomaly detection based on DGBi-SA model,” Engineer- ing Letters, vol. 33, no. 3, pp. 612–619, 2025.
[3]A. M. Idrees, N. S. Elhusseny, and S. Ouf, “Credit card fraud detection model-based machine learning algo- rithms,” IAENG International Journal of Computer Sci- ence, vol. 51, no. 10, pp. 1649–1662, 2024.
[4]A. Alrawajfi, M. T. Ismail, S. Al Wadi, S. Atiewi, and
A. Hamad, “Deep learning approach based on Bi-LSTM for handling missing data in the stock market,” IAENG International Journal of Computer Science, vol. 52, no. 6, pp. 1648–1663, 2025.
[5]Y. Liu and Z. Yao, “Deep learning-based multidi- mensional assessment for network security situations,” IAENG International Journal of Computer Science, vol. 52, no. 9, pp. 3056–3066, 2025.
[6]X. Zhu, W. Cui, Y. Chen, Y. Tao, and X. Wang, “Large language model based on semantic compression for log anomaly detection,” IAENG International Journal of Computer Science, vol. 52, no. 9, pp. 3227–3236, 2025.
[7]J. R. Babu, K. S. Mani, M. Ravindra, R. V. S. L. Ku- mari, R. K. Paladugu, C. R. Babu, and B. Manas, “Hy- brid methods for identifying false data injection attacks in automatic generation control mechanisms,” IAENG In- ternational Journal of Computer Science, vol. 52, no. 9,
pp. 3286–3296, 2025.
[8]Y. Sahin and E. Duman, “Detecting credit card fraud by decision trees and support vector machines,” in Proc. In- ternational MultiConference of Engineers and Computer Scientists (IMECS), vol. I, 2011.
[9]D. Sawh, K. Ponnambalam, and F. Karray, “Artificial in- telligence modeling of financial profit and fraud,” IAENG International Journal of Computer Science, 2011.
[10]J. Akhilomen, “Data mining application for cyber credit- card fraud detection system,” in Proc. World Congress on Engineering, vol. I, 2013.
[11]C. Walgampaya, M. Kantardzic, and R. V. Yampolskiy, “Real time click fraud prevention using multi-level data fusion,” in Proc. World Congress on Engineering and Computer Science (WCECS), vol. I, pp. 514–519, 2010.
[12]P. Shukla et al., “Financial fraud detection and com- parison using different machine learning techniques,” in Proc. 2023 3rd Int. Conf. Technological Advancements in Computational Sciences (ICTACS), IEEE, 2023, pp. 1205–1210.
[13]S. R. Bogireddy and H. Murari, “A hyperparameters tunned ML algorithm for fraud identification in banking and financial transactions,” Int. J. Innovative Science and Research Technology, vol. 9, no. 8, pp. 1619–1625, 2024.
[14]N. K. Trivedi et al., “Credit card fraud detection using machine learning,” Int. J. Scientific Research in Com- puter Science, 2019. [15]M. Rout et al., “Credit card fraud detection using ensem- ble methods,” 2024.
[16]Q. Bao et al., “A deep learning approach to anomaly de- tection in high-frequency trading data,” Int. J. Advanced Computer Science and Applications, 2024.
[17]G. Cao et al., “Real-time anomaly detection in dark pool trading using enhanced transformer networks,” J. Knowl- edge Learning and Science Technology, vol. 3, no. 4, pp. 320–329, 2024.
[18]B. Stojanovic´ et al., “Follow the trail: Machine learning for fraud detection in fintech applications,” Sensors, vol. 21, no. 5, p. 1594, 2021.
[19]G. Jajoo, P. A. Chitale, and S. Aggarwal, “MASCA: LLM based-multi agents system for credit assessment,” arXiv preprint arXiv:2507.22758, 2025.
[20]C.-P. Tsai et al., “AnoLLM: Large language models for tabular anomaly detection,” in Proc. Int. Conf. Learning Representations (ICLR), 2025.
[21]A. Bakumenko et al., “Advancing anomaly detection: Non-semantic financial data encoding with LLMs,” in Proc. Int. Conf. Machine Learning and Applications, 2024.
[22]A. F. Mhammad et al., “Generative & responsible AI – LLMs use in differential governance,” in Proc. 2023 Int. Conf. Computational Science and Computational Intelli- gence (CSCI), IEEE, 2023, pp. 291–296.
[23]S. Korkanti, “Enhancing financial fraud detection using LLMs and advanced data analytics,” in Proc. 2024 2nd Int. Conf. Self Sustainable Artificial Intelligence Frame- works (ICSSAF), IEEE, 2024, pp. 1–6.
[24]G. Singh, P. Singh, and M. Singh, “Advanced real-time fraud detection using RAG-based LLMs,” arXiv preprint arXiv:2501.15290, 2025.
[25]R. Abi, “AI-driven fraud detection systems in fintech us- ing hybrid supervised and unsupervised learning archi- tectures,” Int. J. Research Publication and Reviews, vol. 6, no. 6, pp. 4375–4394, 2025.
[26]A. Narayanan, “Real-time fraud detection in cloud-native fintech systems: A scalable approach using AI and stream processing,” Global J. Engineering and Technology Ad- vances, vol. 23, no. 1, pp. 410–419, 2025.
[27]S. M. Darwish, A. I. Salama, and A. A. Elzoghabi, “In- telligent approach to detecting online fraudulent trading with solution for imbalanced data in fintech forensics,” Scientific Reports, vol. 15, p. 17983, 2025.
[28]F. Xu et al., “Few-shot message-enhanced contrastive learning for graph anomaly detection,” arXiv preprint arXiv:2311.10370, 2023.
[29]
F. Jubiter, “Blockchain and machine learning integra- tion for real-time fraud detection in fintech,” in Proc. Int. Conf. Computer Science, AI, and Machine Learning (ICCSAIML’25), 2025.
[30]S. Bhatnagar, “Leveraging microservices for fraud detec- tion and prevention in fintech: An AI-driven perspective,” Int. Research J. Engineering and Technology (IRJET), vol. 12, no. 6, pp. 1000–1005, 2025.
[31]A. U. Usman et al., “Financial fraud detection using Value-at-Risk with machine learning in skewed data,” IEEE Access, vol. 12, pp. 64285–64296, 2024.
[32]T. Park, “Enhancing anomaly detection in financial mar- kets with an LLM-based multi-agent framework,” arXiv preprint arXiv:2403.19735, 2024.
[33]C. J. Malingu et al., “Application of LLMs to fraud de- tection,” World J. Advanced Research and Reviews, vol. 26, no. 2, pp. 178–183, 2025.
How to Cite This Paper
Vineet Kumar, Sameer Shaik (2025). Large Language Models for Financial Fraud Detection: A Systematic Review of Methodologies, Performance, and Future Directions. International Journal of Computer Techniques, 12(6). ISSN: 2394-2231.
