Abstract
Financial transaction processing systems, for example in payroll, can face challenges where validation is reactive, leading to the late discovery of complex discrepancies after payments are executed. A system and method are described for pre-execution validation of financial transactions. The system can function as a validation layer that applies business rules to batches of transaction data using techniques such as set-based processing. This approach can identify potential anomalies across large datasets before execution. For detected anomalies, a generative model can produce a human-readable explanation detailing a potential rule violation and the associated data. This pre-execution identification and contextualization of potential errors can help reduce the need for post-payment corrections and can improve the efficiency and auditability of large-scale transaction processing.
Creative Commons License
This work is licensed under a Creative Commons Attribution 4.0 License.
Recommended Citation
Jha, Shalini; Rawani, Farhan; Mangalampalli, Bhaskar; Campus, Alba; Seerapu, Sanhitha; Martin, Millie; Shan, Arib; Basu, Tilottama; Badugu, Amogh; Krishna, Sai; and Karyampudi, Ramya, "Rule Based Pre-Execution Validation of Financial Transactions with Generative Anomaly Explanations", Technical Disclosure Commons, (September 16, 2026)
https://www.tdcommons.org/dpubs_series/11736