SAGE: An LLM-driven Self Reflective Agentic Framework for Fraud Detection

• URL: http://arxiv.org/abs/2606.08146v1
• Date: Sat, 06 Jun 2026 12:48:45 GMT
• Title: SAGE: An LLM-driven Self Reflective Agentic Framework for Fraud Detection
• Abstract summary: We propose SAGE, the first end-to-end multi-agent framework for fraud detection.<n>SAGE coordinates three dedicated agents that make decisions based on a six-layer Data Diagnostic Tree (DDT) and a Markov decision process guided by natural-language gradients.<n>On five fraud datasets and five backbone LLMs, SAGE wins $96.00%$ of method--dataset comparisons and improves F1 by an average of $40.86%$ over baselines.
• Score: 19.81280139275702
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Fraud detection in payment, e-commerce, and telecommunications systems requires accuracy at the individual level, robustness under severe class imbalance, and ease of understanding for risk managers. Existing methods fall at least one of these requirements: automated machine learning systems search a fixed numerical space without semantic awareness of the dataset; graph neural network-based methods require pre-defined relational graphs and remain opaque at the individual-decision level; and the design of general-purpose large language model (LLM) agents does not consider the recall and precision constraints specific to real-world fraud detection. In this paper, we propose SAGE, the first end-to-end LLM-driven multi-agent framework for fraud detection. SAGE coordinates three dedicated agents that make decisions based on a six-layer Data Diagnostic Tree (DDT) and a Markov decision process guided by natural-language gradients, automatically optimizing the model under a fraud-specific reward. On five fraud datasets and five LLM backbones, SAGE wins $96.00\%$ of method--dataset comparisons and improves F1 by an average of $40.86\%$ over baselines. The code is available at https://github.com/yichenC1c/SAGE.

Related papers

• Knowledge Over Parameters: Evolving Smart Contract Vulnerability Detection [14.891723124328822]
  We present EvoVuln, an automated framework that reformulates vulnerability detection as a procedural knowledge evolution problem.<n>We evaluate it across five real-world vulnerability types and find it to be 71% a macro-average F1-score, outperforming all baselines.
  arXiv  Detail & Related papers  (2026-07-02T05:54:23Z)
• Trustworthy AI: Ensuring Reliability and Accountability from Models to Agents [2.3407932351117324]
  We develop algorithms with theoretical guarantees for ensuring reliability and accountability of Machine Learning (ML) systems.<n>This thesis introduces tools grounded in information theory, optimization, and statistical learning to mitigate bias, reduce arbitrary decisions, and ensure content provenance.<n>We ensure accountability in generative AI through watermarking large language models (LLMs)
  arXiv  Detail & Related papers  (2026-05-09T14:11:36Z)
• Reliable LLM-Based Edge-Cloud-Expert Cascades for Telecom Knowledge Systems [54.916243942641444]
  Large language models (LLMs) are emerging as key enablers of automation in domains such as telecommunications.<n>We study an edge-cloud-expert cascaded LLM-based knowledge system that supports decision-making through a question-and-answer pipeline.
  arXiv  Detail & Related papers  (2025-12-23T03:10:09Z)
• Check Field Detection Agent (CFD-Agent) using Multimodal Large Language and Vision Language Models [7.836288735110501]
  We introduce a novel, training-free framework for automated check field detection.<n>Our approach enables zero-shot detection of check components, significantly lowering the barrier to deployment in real-world financial settings.
  arXiv  Detail & Related papers  (2025-09-22T20:43:59Z)
• SciML Agents: Write the Solver, Not the Solution [69.5021018644143]
  We introduce two new datasets: a diagnostic dataset of adversarial "misleading" problems; and a large-scale benchmark of 1,000 diverse ODE tasks.<n>We evaluate open- and closed-source LLM models along two axes: (i) unguided versus guided prompting with domain-specific knowledge; and (ii) off-the-shelf versus fine-tuned variants.<n>Preliminary results indicate that careful prompting and fine-tuning can yield a specialized LLM agent capable of reliably solving simple ODE problems.
  arXiv  Detail & Related papers  (2025-09-12T02:53:57Z)
• MCP-Orchestrated Multi-Agent System for Automated Disinformation Detection [84.75972919995398]
  This paper presents a multi-agent system that uses relation extraction to detect disinformation in news articles.<n>The proposed Agentic AI system combines four agents: (i) a machine learning agent (logistic regression), (ii) a Wikipedia knowledge check agent, and (iv) a web-scraped data analyzer.<n>Results demonstrate that the multi-agent ensemble achieves 95.3% accuracy with an F1 score of 0.964, significantly outperforming individual agents and traditional approaches.
  arXiv  Detail & Related papers  (2025-08-13T19:14:48Z)
• Which Agent Causes Task Failures and When? On Automated Failure Attribution of LLM Multi-Agent Systems [50.29939179830491]
  Failure attribution in LLM multi-agent systems remains underexplored and labor-intensive.<n>We develop and evaluate three automated failure attribution methods, summarizing their corresponding pros and cons.<n>The best method achieves 53.5% accuracy in identifying failure-responsible agents but only 14.2% in pinpointing failure steps.
  arXiv  Detail & Related papers  (2025-04-30T23:09:44Z)
• Explainable AI for Fraud Detection: An Attention-Based Ensemble of CNNs, GNNs, and A Confidence-Driven Gating Mechanism [5.486205584465161]
  This study presents a new stacking-based approach for CCF detection by adding two extra layers to the usual classification process.<n>In the attention layer, we combine soft outputs from a convolutional neural network (CNN) and a recurrent neural network (RNN) using the dependent ordered weighted averaging (DOWA) operator.<n>In the confidence-based layer, we select whichever aggregate (DOWA or IOWA) shows lower uncertainty to feed into a meta-learner.<n>Experiments on three datasets show that our method achieves high accuracy and robust generalization, making it effective for CCF detection.
  arXiv  Detail & Related papers  (2024-10-01T09:56:23Z)
• Advancing Anomaly Detection: Non-Semantic Financial Data Encoding with LLMs [49.57641083688934]
  We introduce a novel approach to anomaly detection in financial data using Large Language Models (LLMs) embeddings. Our experiments demonstrate that LLMs contribute valuable information to anomaly detection as our models outperform the baselines.
  arXiv  Detail & Related papers  (2024-06-05T20:19:09Z)
• Securing Transactions: A Hybrid Dependable Ensemble Machine Learning Model using IHT-LR and Grid Search [2.4374097382908477]
  We introduce a state-of-the-art hybrid ensemble (ENS) Machine learning (ML) model that intelligently combines multiple algorithms to enhance fraud identification. Our experiments are conducted on a publicly available credit card dataset comprising 284,807 transactions. The proposed model achieves impressive accuracy rates of 99.66%, 99.73%, 98.56%, and 99.79%, and a perfect 100% for the DT, RF, KNN, and ENS models, respectively.
  arXiv  Detail & Related papers  (2024-02-22T09:01:42Z)
• Explainable Fraud Detection with Deep Symbolic Classification [4.1205832766381985]
  We present Deep Classification, an extension of the Deep Symbolic Regression framework to classification problems. Because the functions are mathematical expressions that are in closed-form and concise, the model is inherently explainable both at the level of a single classification decision and the model's decision process. An evaluation on the PaySim data set demonstrates competitive predictive performance with state-of-the-art models, while surpassing them in terms of explainability.
  arXiv  Detail & Related papers  (2023-12-01T13:50:55Z)
• Fraud Dataset Benchmark and Applications [25.184342958800293]
  Fraud dataset Benchmark (FDB) is a compilation of publicly available datasets catered to fraud detection. FDB comprises variety of fraud related tasks, ranging from identifying fraudulent card-not-present transactions, detecting bot attacks, classifying malicious URLs, estimating risk of loan default to content moderation. Python based library for FDB provides a consistent API for data loading with standardized training and testing splits.
  arXiv  Detail & Related papers  (2022-08-30T17:35:39Z)
• Coded Stochastic ADMM for Decentralized Consensus Optimization with Edge Computing [113.52575069030192]
  Big data, including applications with high security requirements, are often collected and stored on multiple heterogeneous devices, such as mobile devices, drones and vehicles. Due to the limitations of communication costs and security requirements, it is of paramount importance to extract information in a decentralized manner instead of aggregating data to a fusion center. We consider the problem of learning model parameters in a multi-agent system with data locally processed via distributed edge nodes. A class of mini-batch alternating direction method of multipliers (ADMM) algorithms is explored to develop the distributed learning model.
  arXiv  Detail & Related papers  (2020-10-02T10:41:59Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.