Source record

Published by arXiv cs.HC (T1 source). The original is at https://arxiv.org/abs/2607.10317.

Pipeline notes

The summary and note below are generated by the signal pipeline — they are Beyond Desk’s reading, not quotations from the source.

Research paper proposing a hierarchical human-AI triage model for POS fraud detection in Nigerian FinTech, using a calibrated ensemble filter, three-tier routing, dynamic shadow pricing to ration human attention, and random audits. Reports a regional performance gap reduction from 19.43 to 2.88 percentage points and 24.79-point gain in fraud recall over an autonomous baseline.

The three-tier triage with shadow-priced human attention and audit-based skill maintenance is a transferrable design pattern, and the analysis of infrastructure noise as false fraud signal is broadly applicable.

Cited by

No citations on record.