# Fraud Type Decomposition and the Observation-Mechanism Taxonomy:Class-Specific Detection Limits in Payment Networks

arXiv: https://arxiv.org/abs/2605.31257 · HTML: https://mlsanity.com/paper/2605.31257/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 2 of its 12 references that are in the map.
- First posted on arXiv 29 May 2026.

## Builds on (most cited first, 2 of 2)

- [The Fundamental Limits of Fraud Detection in Card Payment Networks](https://mlsanity.com/paper/2605.27557/) (2026)
- [Causal Label Recovery in Payment Networks](https://mlsanity.com/paper/2605.29272/) (2026)

## Abstract (arXiv)

Fraud detection in payment networks relies on labels generated through heterogeneous and imperfect observation processes, yet existing approaches treat fraud as a homogeneous binary variable. We show that this assumption is structurally incorrect and leads to provable inefficiency. We introduce an observation-mechanism taxonomy that partitions fraud into five classes, each defined by a distinct censorship and labeling pipeline. We prove that estimating fraud rates separately by class and aggregating strictly dominates pooled estimation, with the efficiency gap characterized as a Jensen penalty arising from heterogeneous observation rates. For each class, we derive the binding theoretical constraint on detection, including endogenous label corruption, structural non-observability, and feature non-informativeness. These results establish that fraud detection is fundamentally a collection of distinct estimation problems, each governed by its own observation structure and detection limit.