Quantum-aware market manipulation surveillance

Manipulation hides in the web of who trades with whom — not in the price.

Spoofing, wash trading and collusion leave a high-order relational footprintspread across accounts, instruments and venues — rare, deliberately buried, and invisible to per-account rules (even the banks' own surveillance misses it — see JP Morgan).

We turn every surveillance window into that relational structure, surface the coordinated patterns — cycles, cliques, layering — rank them by calibrated confidence, and measure exactly when a quantum sketch is needed to hold it (an effective-rank criterion).

surveillance feed · live (demo)streaming
ingesting order flow

order flow → relational view → coordinated motif surfaced

01The problem

Rules and classical thresholds watch isolated accounts. Modern manipulation is higher-order.

This is an anomaly-detection system. The residue where abuse lives is the order flow and a relational structure between traders, across time and venues. Three typologies carry most of it — and spoofing is the headline.

spoofing · headline typology

Layering & cancel

Spoofing is placing large orders with no intent to execute — false price signals — then cancelling before execution and trading the other side. The intent never rests in the trades; it lives in the place-and-cancel pattern of the order flow.

Coordinated, cross-account spoofing splits the layering across colluding accounts — a higher-order relational signature that per-account rules and 1-WL GNNs can miss.

BIDASKdecoyreal fillmidillustrative · layering & cancel → execute opposite side
wash trading

Cycles

A group buys and sells among themselves in a loop. Per account the volume looks ordinary; the signal is the cycle, not the size.

collusion

Cliques

Several accounts act in lockstep as a dense block. Taken one by one, none of them raises a flag.

Capturing this structure classically is costly — but not always as costly as it looks. That is exactly the question we measure.

02Why it's hard

The structure hides in higher-order motifs — and the classical tools are bounded.

motivating case · JPMorgan

$0.0M

2020 CFTC penalty — largest ever, a record for spoofing

Over roughly eight years (2008–2016), 15 traders across two desks placed hundreds of thousands of spoof orders in precious-metals and U.S. Treasury futures — more than $300M in harm to other market participants.

The kicker: in 2024 the OCC and the Federal Reserve added $250M and $98.2M penalties for trade-surveillance failures — their own monitoring didn't catch it.

Spoofing is illegal, enormous, and even tier-1 surveillance misses it.

source: CFTC press release 8260-20 ↗
01

Rules & thresholds

Most surveillance today is rules and thresholds. It drowns analysts in false positives while missing the subtle, coordinated schemes — the alerts that fire are rarely the ones that matter, and the ones that matter rarely fire.

02

The GNN expressivity ceiling

Standard message-passing GNNs are bounded by the 1-Weisfeiler-Leman test. They provably cannot reliably count or distinguish the higher-order motifs where coordinated manipulation lives — cycles (wash trading), cliques (collusion), and cross-account layering (coordinated spoofing). Capturing that higher-order structure exactly is ~N^k classically — the real wall.

03

Where the quantum sketch fits

A quantum oracle sketch is an embedding of exponential capacity held in polylog space. It can carry a high-order relational representation that is intractable to store classically — exactly the regime the motifs above demand.

the quantum advantage, honestly

The QOS space advantage is exponential and proven without hardness conjectures — unlike time-speedup methods, which have largely been dequantized. It holds a 2^d-dimensional relational representation in about d qubits.

It is a space advantage, not speed, and it only bites when the manipulation residual is genuinely high effective rank.

We don't claim quantum wins today — at simulable scales a classical embedding suffices. We give the falsifiable criterion — the effective rank of the manipulation residual — and the instrument to measure when the quantum regime is reached.

03Feature construction

Fifty base signals, multiplied in pairs, packed into one vector.

This is purely the mechanism — how the feature vector is assembled, nothing about detection or results. Fifty base features are z-scored, then every pair is multiplied to form order-2 correlations, expanding 50 → 1,275 before the vector is amplitude-encoded.

50 BASE FEATURESz-scored — mean 0, std 1, so every feature weighs the same before they combineNETWORK · 23ORDER FLOW · 22VOLUME · 5ORDER-2 CORRELATION — A PAIR PRODUCTotr_maxHIGHimbalance_volatility×otr_max × imbalance_volatilityfires only when BOTH are highINTENSITY LOGIChigh × low → lowlow × high → lowhigh × high → HIGHAn order-2 correlation = the product of two z-scored features.EXPANSION — EVERY PAIRconnect all pairs of the 50 features (upper triangle)FEATURES501,27550 + 50·49/2 = 1,225 → 1,275 featuresbase + all pairwise productsPACKAGED VECTOR1,275 values50 base features + their order-2 products, laid out as one vector.2¹¹ = 2,048 amplitudes → 13 qubits (11 + 2)order-3 → ~2¹⁵
Hover a chip to read its name · click two chips to form their order-2 product
expansion order
04The method

One head. Two representations. The only variable is the embedding.

One surveillance window, two embeddings — a QOS quantum sketch and a GNN — feeding the sameridge / LS-SVM head. Hold the head byte-for-byte fixed and the only variable left is the representation, so any difference is the embedding's. What decides whether the quantum sketch is needed is the effective rank of the manipulation residual.

The circuit, the GNN and the full pipeline live next door.
see the full circuit & pipeline ↓

surveillancewindowsame dataQOS sketchrepresentation AGNN propagationrepresentation Bonly the embedding differsRidge head(LS-SVM)byte-for-byte identicalanomaly score→ ranked alerts
04.1quantum advantage · machine-size scaling
why look at more correlations

One account at a time, manipulation is invisible: each trader's volume and timing look ordinary. The signal lives in the relations between accounts, not in any single one.

Order-2 features encode who traded with whom — the co-trading matrix that exposes wash-trading cycles and collusion cliques a per-account view never sees. Each higher order of correlation captures more intricate coordination, and that is precisely what surfaces schemes invisible at lower order.

The catch is combinatorial: pairwise features scale as ~N², order-k as ~N^k. More correlations mean more information — and a feature dimension that explodes. That growth is the point, not a flaw: it is the only way to make these anomalies detectable. What it costs is space — which is exactly what the chart below measures, classical versus quantum, and where the classical machine stops being viable.

classical vs quantum machine sizelog scale · drag b
10¹101010¹³10¹⁷8152030405060feature bits b (d = 2ᵇ) →classical storage limit (~10¹⁵)classical storage infeasible →classical — 2ᵇ floatsquantum — 2b+18 qubitsb = 15

The advantage is space, not detection. At every simulable scale both representations detect equally well (AUC 0.998 vs 1.000). It becomes real only where classical storage fails (d ≈ 2⁵⁰) AND the effective rank is genuinely high.

04.3what we actually measured
detection · 2¹³ features · recall

Detection parity

0.00.51.0recall →GNN0.916QOS0.854

At every scale we can simulate, detection ties — the classical side even leads. The advantage was never precision.

run on real hardware · iqm emerald

Real hardware · 4 qubits · 2⁴ features

simulation (outline)hardware (fill)0.99FIDELITY VSSIMULATION0000000100100011010001010110011110001001101010111100110111101111output state (2⁴) →

The measured distribution reproduces simulation at 0.99 fidelity — this validates the circuit on real hardware, not a quantum edge.

05Inside the sketch

From a 2⁶⁰-dimensional feature to the ridge head — and the circuit that carries it.

Relational featurev ∈ ℝ^(2^60)2^60 floatsAmplitude-encodeinto ~60data qubits~60 qubitsQOS state-sketchPrep·O_h·U_data·LCU₁QSVT·LCU₂·Frame⁻¹·postselect~60 + 2 ancillasReadoutmeasure /inner productscost: measurement samplesRidge / LS-SVM headsame head asthe GNN pathlow-dimCalibrated score→ rankedalertscalibratedclassical 2^60 floats ≈ 1.15×10^18 (beyond any storage) vs sketch ~60 qubits + 2 ancillas + overhead (∝ log₂ dim)2^60 illustrates the regime where classical storage breaks
QOS state-sketch

The q_state_sketch circuit: equal superposition (Prep), random ±1 frame (O_h), phase oracle averaged over samples (U_data), phase→sin via LCU₁, arcsin synthesis via QSVT, real-part extraction (LCU₂), inverse frame (Frame⁻¹), then postselect a1 = a2 = |0⟩. The data register holds a low-variance state-sketch.

Open the circuit playground ↓
honest caveat

Space advantage, not speed; readout costs measurement samples; the advantage only bites when the manipulation residual is genuinely high effective rank. At simulable scales a classical embedding suffices — we give the falsifiable criterion that says when the quantum regime is reached.

05.1circuit playground

The faithful q_state_sketch circuit for N = 2ⁿ. Slide n to grow the data register, and click any composite oracle to open its real internal subcircuit.

dim = 22 = 4qubits ≈ 2 data + 2 ancillasclassical: 4 floats · quantum: ⌈log₂dim⌉ + 2 (+overhead)
a2a1q1q0sketchHHHO_hO_hU_dataU_dataLCU₁+QSVTLCU₁+QSVTLCU₂LCU₂Frame⁻¹Frame⁻¹M
Prep / framePhase oracleLCUQSVTReal partMeasure
q_state_sketch

Encodes a 2n-dimensional relational feature into n data qubits + 2 ancillas, then postselects a1 = a2 = |0⟩. Click a composite gate (marked ⊕) to open its real subcircuit from the QOS paper.

06A day as an analyst

The alert queue, prioritized. Pick one and see why.

surveillance-console · live feed (demo)
alert queueranked by confidence
BIDASKdecoyreal fillmidillustrative · layering & cancel → execute opposite side
window

s-0003

verdict

Spoofing (layering & cancel)

patternlayering
actionplace + cancel
subtlety0.12
order≥ 2

The decoy ladder posts and cancels to move the mid, then a real order executes on the opposite side. The alert fires on the place-and-cancel pattern, not on any single resting order.

07Who we are

Team Imperivm — at the intersection of quantum, ML and markets.

Universidad Autónoma de Madrid (UAM), built at the Junction Quantum Hackathon. Quantum singular-value transformation, ML applied to fraud detection, and market-microstructure expertise — we build the instrument that measures the question, not the hype around it.

Enrique Anguiano-Vara

CSIC Researcher

Connect on LinkedIn ↗

Ignacio López Leis

MSc Quantum Computing (UAM)

Connect on LinkedIn ↗

Manuel Esparcia Cantos

MSc Quantum Computing (UAM)

Connect on LinkedIn ↗
The criterion

The quantum primitive is real and its space advantage exists. What we measure is when a surveillance task actually needs it.

A falsifiable effective-rank criterion and an end-to-end pipeline to measure it.

arXiv:2604.07639 · QOS (Zhao, Preskill et al.) · demo with synthetic data

Watch · 2-minute overview