A Cognitive-Aware QML-CRL Framework for Detecting Affinity and Romance-Investment Fraud

Bibhas Adhikari; Ramya Srinivasan

cs.LGOct 6, 2026

A Cognitive-Aware QML-CRL Framework for Detecting Affinity and Romance-Investment Fraud

Authors: Bibhas Adhikari, Ramya Srinivasan

Organizations: Fujitsu Research of America, Inc., Santa Clara, CA, USA · Security Science Lab, Fujitsu Research of America, Santa Clara, CA 95054 Inc.

Abstract

We present a hybrid quantum-classical framework that detects affinity and romance-investment fraud by modelling the cognitive biases in a manipulative conversation. In our proposed framework, cognitive biases central to this fraud class are carried by dedicated qubits in a structured parameterized quantum circuit, together with a frame qubit makes the encoding sensitive to the temporal order of manipulative reframing, and a narrative qubit that aggregates co-occurrence through a trainable entanglement layer. The circuit parameters are trained jointly with a classical reinforcement-learning agent that decides, turn by turn, whether to flag the conversation, modeled as an optimal stopping problem. We evaluate the model's performance on synthetic conversations that include hard negatives, legitimate but urgent, and legitimate but pushy sales conversations.

Figures & tables

Panel of Figure 1 — Figure 1 : Complete architecture. (a) The end-to-end pipeline from raw conversation text to a flag/wait decision. (b) Detail of the parameterized quantum-cognitive encoder.

Panel of Figure 2 — Figure 2 : Gate-level structure of the parameterized quantum-cognitive encoder. All angles shown are functions of the trainable parameters $\Theta$ and of the current turn’s cues.

Stage	loss	commit	recip.	auth.	frame
Contact	0.05	0.05	0.05	0.10	0.0
Rapport	0.10	0.25	0.10	0.20	0.0
Reciprocity hook	0.10	0.30	0.65	0.30	0.0
Authority + big ask	0.20	0.45	0.30	0.60	0.0
Loss event	0.70	0.55	0.15	0.40	$-0.8$
Recovery pressure	0.65	0.75	0.20	0.55	$+0.8$

Table 1 : Fraud class base cue profile

b^{(s)}

by stage (Eq. 11 ).

Configuration	$F_{1}$	FPR	Lateness	FN
PQC, joint, cognitive init	$0.934\pm 0.019$	$0.000$	$+2.19$	$7.7$
PQC, frozen, cognitive init	$0.919\pm 0.051$	$0.000$	$+1.68$	$9.0$
Classical cue baseline	$0.899\pm 0.020$	$0.000$	$+0.99$	$11.3$
PQC, frozen, mis-specified init	$0.000\pm 0.000$	$0.000$	—	$62.0$
PQC, joint, mis-specified init	$\mathbf{0.971\pm 0.041}$	$0.000$	$+1.37$	$3.3$
PQC, joint, no frame qubit	$0.000\pm 0.000$	$0.000$	—	$62.0$

Table 2 : Detection quality on the hard synthetic benchmark (300 test conversations, 20% fraud base rate, mean

\pm

s.d. over three seeds). Lateness is the mean flag turn minus the labelled critical loss turn, over true positives; it is undefined for configurations that never flag.

Panel of Figure 3 — Figure 3 : Training and order-sensitivity results. (a) Joint training enables the mis-specified encoder to adapt after warm-up, whereas the frozen encoder fails to flag. (b) Detection timing under turn-order shuffling with cue content held fixed.

Figure 4 : Risk-feature trajectories at the cognitive initialization (top) and after joint training (bottom), for a fraud conversation, a pushy-sales hard negative, and a benign conversation. Training sharpens the frame channel into the discriminator that separates fraud from pushy sales, while the loss and authority channels saturate on all three classes.

Episode (one labelled conversation)
Symbol	Meaning
$T$	number of turns
$c_{t}\in[0,1]^{n}\!\times\![-1,1]$	cue vector at turn $t$ from the NLP front end
$y\in\{0,1\}$	conversation label; $1$ = fraud
$t^{\ast}$	labelled critical loss turn (first turn of the loss stage)
Quantum encoder

Table 3 : Notation used in Algorithm 2 .

Figure 5 : The synthetic conversation generator. (a) Each fraudulent conversation is divided into six stages by the boundaries $\lfloor\mathrm{linspace}(0,T,7)\rfloor$ , shown here for $T=14$ ; the labelled critical turn $t^{\ast}$ is the first turn of the loss-event stage (turn 9), and detection lateness is measured relative to it. (b) Each stage $s$ carries a fixed base level $b^{(s)}_{i}\in[0,1]$ for the four bias channels. (c) The frame channel $b^{(s)}_{f}\in[-1,1]$ is zero for the first four stages and non-zero only in the last two, with opposite signs, so every fraudulent conversation contains exactly one $R_{Z}\!\leftrightarrow\!R_{Y}$ transition. By the sign convention of Eq. ( 6 ), $c_{t,f}>0$ is loss framing and drives $R_{Z}$ while $c_{t,f}<0$ is recovery framing and drives $R_{Y}.$ (d,e) Sampled realizations, drawn as $c_{(t,i)}=\mathrm{clip}(b^{(s)}_{i}+\varepsilon,\ \ell_{i},u_{i})$ with $\varepsilon\sim\mathcal{N}(0,\sigma^{2})$ , $\sigma=0.10$ , independently per turn and per channel, so there is within-stage variation but no autoregressive structure. Shading marks the stage, the dashed red line marks $t^{\ast}$ .

Explore similar work

Q-SYNTH: Hybrid Quantum-Classical Adversarial Augmentation for Imbalanced Fraud Detection
May 20, 2026Adam Innan, Mansour El Alami, Nouhaila Innan +2Hybrid Quantum-Classical Pipeline Fraud Detection
Quantum Physics and Spintronics Team, LPMC, Faculty of Sciences Ben M’sick, Hassan II University of Casablanca, Morocco · eBRAIN Lab, Division of Engineering, New York University Abu Dhabi (NYUAD), Abu Dhabi, UAE · Center for Quantum and Topological Systems (CQTS), NYUAD Research Institute, NYUAD, Abu Dhabi, UAE
When Similarity Is Interaction-Driven: Quantum Kernels for Regime-Sensitive Learning
Aug 25, 2026Hanqiu Peng, Jianlong Lu, Ying ChenQuantum Kernels Kernel Method
Centre for Quantitative Finance, Department of Mathematics Risk Management Institute, National University of Singapore, Singapore
Repair Before Veto, When Repair Is Hidden: Quantum-Accessible Features for Repair-Augmented Constraint Learning
Jun 6, 2026Yifan WangQuantum Learning Quantum Error Correction
Department of Mechanical Engineering, McGill University, QC, H3A 2T7, Canada

May 20, 2026, cs.LGJ/K move · Enter open · S save

Q-SYNTH: Hybrid Quantum-Classical Adversarial Augmentation for Imbalanced Fraud Detection

Adam Innan, Mansour El Alami, Nouhaila Innan, Muhammad Shafique+1

Quantum Physics and Spintronics Team, LPMC, Faculty of Sciences Ben M’sick, Hassan II University of Casablanca, Morocco · eBRAIN Lab, Division of Engineering, New York University Abu Dhabi (NYUAD), Abu Dhabi, UAE · Center for Quantum and Topological Systems (CQTS), NYUAD Research Institute, NYUAD, Abu Dhabi, UAE

Hybrid Quantum-Classical Pipeline Fraud Detection Oversampling Generative Adversarial Network

Open paper AskPDF arXiv

Credit card fraud detection is fundamentally challenged by extreme class imbalance, where fraudulent transactions are rare yet operationally critical. This imbalance often biases supervised learners toward the legitimate class, leading to high overall accuracy but weaker fraud-class recall and F1-score. This paper introduces Q-SYNTH, a hybrid classical--quantum generative adversarial framework in which a parameterized quantum circuit serves as the generator and a classical neural network serves as the discriminator. Q-SYNTH is designed for minority-class fraud synthesis in tabular data and is evaluated along two dimensions: statistical fidelity to real fraud samples and downstream performance for fraud detection. To this end, generated samples are assessed using distributional similarity measures based on Kolmogorov-Smirnov statistics and Wasserstein distances, real-vs-synthetic detectability measured by AUC-ROC, and downstream classification performance across both quantum and classical classifiers. Under the reported protocol, Q-SYNTH reduces marginal distribution mismatch relative to a classical GAN baseline while maintaining competitive downstream fraud-detection performance. Although SMOTE achieves the strongest feature-wise similarity and the classical GAN attains the highest downstream performance in several settings, Q-SYNTH offers a favorable compromise between distributional fidelity and downstream performance, supporting the feasibility of hybrid quantum augmentation for imbalanced fraud detection.

Abstract

Explore similar work

Q-SYNTH: Hybrid Quantum-Classical Adversarial Augmentation for Imbalanced Fraud Detection

When Similarity Is Interaction-Driven: Quantum Kernels for Regime-Sensitive Learning

Repair Before Veto, When Repair Is Hidden: Quantum-Accessible Features for Repair-Augmented Constraint Learning