# Unsupervised Graph Modeling for Anomaly Detection in Accounting Subject Relationships

arXiv: https://arxiv.org/abs/2604.26216 · HTML: https://mlsanity.com/paper/2604.26216/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 16 of its 50 references that are in the map.
- First posted on arXiv 29 Apr 2026.

## Builds on (most cited first, 16 of 16)

- [Rethinking Graph Neural Networks for Anomaly Detection](https://mlsanity.com/paper/2205.15508/) (2022)
- [DGraph: A Large-Scale Financial Dataset for Graph Anomaly Detection](https://mlsanity.com/paper/2207.03579/) (2022)
- [Recurrent Neural Network from Adder's Perspective: Carry-lookahead RNN](https://mlsanity.com/paper/2106.12901/) (2021)
- [Trustworthy Summarization via Uncertainty Quantification and Risk Awareness in Large Language Models](https://mlsanity.com/paper/2510.01231/) (2025)
- [AI-Driven Health Monitoring of Distributed Computing Architecture: Insights from XGBoost and SHAP](https://mlsanity.com/paper/2501.14745/) (2024)
- [Structural Generalization for Microservice Routing Using Graph Neural Networks](https://mlsanity.com/paper/2510.15210/) (2025)
- [Parameter-Efficient Fine-Tuning with Differential Privacy for Robust Instruction Adaptation in Large Language Models](https://mlsanity.com/paper/2512.06711/) (2025)
- [ADAMM: Anomaly Detection of Attributed Multi-graphs with Metadata: A Unified Neural Network Approach](https://mlsanity.com/paper/2311.07355/) (2023)
- [Temporal Graph Networks for Graph Anomaly Detection in Financial Networks](https://mlsanity.com/paper/2404.00060/) (2024)
- [Anomaly Detection in Double-entry Bookkeeping Data by Federated Learning System with Non-model Sharing Approach](https://mlsanity.com/paper/2501.12723/) (2025)
- [GADY: Unsupervised Anomaly Detection on Dynamic Graphs](https://mlsanity.com/paper/2310.16376/) (2023)
- [Knowledge-Augmented Large Language Model Agents for Explainable Financial Decision-Making](https://mlsanity.com/paper/2512.09440/) (2025)
- [Dynamic Anomaly Identification in Accounting Transactions via Multi-Head Self-Attention Networks](https://mlsanity.com/paper/2511.12122/) (2025)
- [Generative Modeling of Human-Computer Interfaces with Diffusion Processes and Conditional Control](https://mlsanity.com/paper/2601.06823/) (2026)
- [Graph-Structured Deep Learning Framework for Multi-task Contention Identification with High-dimensional Metrics](https://mlsanity.com/paper/2601.20389/) (2026)
- [CaliCausalRank: Calibrated Multi-Objective Ad Ranking with Robust Counterfactual Utility Optimization](https://mlsanity.com/paper/2602.18786/) (2026)

## Abstract (arXiv)

This paper addresses the problem of anomaly detection in accounting subject association structures, proposing a structured modeling and unsupervised discriminant framework based on graph neural networks. This framework is used to mine stable correspondences between subjects and identify structural deviations from general ledger details and voucher entries. The method first abstracts accounting subjects as graph nodes, and the co-occurrence and debit/credit correspondence of subjects in the same business record are abstracted as weighted edges. The edge weights are characterized by statistical measures such as co-occurrence frequency or amount aggregation, thus forming a period-level accounting subject association graph. In the representation learning stage, a message passing mechanism is used to fuse the node's own attributes and neighborhood context to obtain node embeddings containing structural information. In the anomaly detection stage, the rationality of subject pair connections is estimated through a relation reconstruction decoder, and edge-level anomaly scores are defined based on the degree of deviation in reconstruction probabilities. These scores are then aggregated to obtain node-level risk ranking and local anomaly localization. This framework can simultaneously capture local substructure anomalies and cross-community anomaly connections without relying on anomaly labeling, outputting traceable subject pair risk clues. Comparative experiments demonstrate more stable comprehensive discriminant capabilities and higher top-ranking accuracy.