Testing a continuous-variable Bell-like inequality with a hybrid-encoded system

This paper demonstrates a 380-standard-deviation violation of a Bell-like noncontextual hidden-variable inequality using a hybrid-encoded system that maps single-photon spatial modes to Gottesman–Kitaev–Preskill code space, proving that sequential measurements on continuous-variable systems can reveal quantum noncontextuality where standard quadrature measurements cannot.

The Gist

Imagine you are trying to prove that the universe is fundamentally weird and doesn't follow the rules of a simple, predictable machine. For decades, scientists have used "Bell tests" to show that quantum particles are connected in ways that defy common sense (like two dice rolling the same number instantly, no matter how far apart they are).

However, there's a specific type of quantum weirdness called contextuality that is harder to catch. Think of it like this: In a normal world, if you ask a person, "What is your favorite color?" the answer shouldn't change just because you also asked, "What is your favorite food?" at the same time. But in the quantum world, the answer to "What is your favorite color?" does change depending on what other question you ask alongside it. This is "contextuality."

The Problem with "Continuous" Systems
Most previous experiments proving this used "discrete" systems, like tiny switches that are either ON or OFF (0 or 1). But scientists also want to test this on "continuous" systems, which are more like a dimmer switch that can be set to any value along a smooth line.

The trouble is, measuring these smooth, continuous systems usually destroys the delicate quantum state, like trying to weigh a soap bubble by poking it with a needle. If you poke it, it pops, and you can't see the weird quantum behavior anymore. For a long time, it seemed impossible to prove contextuality in these smooth systems without destroying the evidence.

The New Trick: The "Hadamard Test" as a Shadow Puppet
The team in this paper found a clever workaround. Instead of poking the bubble directly, they used a "shadow puppet" technique.

1. The Setup: They used a single photon (a particle of light) generated by a tiny semiconductor dot. This photon has two "personas":

   • The Control (The Puppeteer): Its polarization (the direction it vibrates) acts like a switch (ON/OFF).
   • The Target (The Bubble): Its position in space acts like the smooth, continuous dimmer switch.

2. The Game: They set up a "Peres-Mermin Square," which is like a 3x3 grid of rules. In a normal, non-quantum world, you can fill this grid with numbers that satisfy all the rules at once. In the quantum world, the rules contradict each other, making it impossible to fill the grid without breaking a rule.

3. The Measurement: Instead of measuring the photon's position directly (which would destroy it), they used a "Hadamard test." Imagine you have a magic mirror. You don't look at the object directly; instead, you look at its reflection in a mirror that has been slightly tilted by the object. By measuring the tilt of the reflection, you can figure out the object's properties without ever touching it.

What They Found
By using this "shadow puppet" method, they were able to check the rules of the 3x3 grid without destroying the photon's quantum state.

• The Result: The numbers they got didn't fit the "normal world" rules at all. They violated the inequality (the rulebook for non-quantum reality) by a massive margin—380 standard deviations. To put that in perspective, if you flipped a coin 380 times and it landed on heads every single time, that would be a statistical miracle. This result is that kind of miracle.

Why It Matters
This experiment is a big deal because:

• It's a "Black Box" Test: They didn't need to assume the quantum theory was correct to prove it. They just put the system in a box, ran the test, and the result spoke for itself.
• It Works on Smooth Systems: They proved that you can see this deep quantum weirdness in continuous systems, not just in simple ON/OFF switches.
• No "Popping": They managed to do this without destroying the quantum state, which was the biggest hurdle before.

The Catch
The paper does admit one small imperfection: the "puppets" (the optical tools they used) weren't perfectly synchronized. There was a tiny bit of "jitter" in how the rules were applied. However, the violation was so huge that even with this jitter, the quantum weirdness was undeniable. They couldn't mathematically fix the jitter to make the proof "perfect," but the evidence is strong enough to say, "Yes, the universe is contextually weird, even in these smooth systems."

In short, they built a clever, non-destructive way to peek behind the curtain of reality and confirmed that the universe plays by rules that are far stranger than our everyday experience suggests.

Technical Summary

1. Problem Statement

The paper addresses a fundamental challenge in quantum foundations: testing contextuality (a form of nonclassicality broader than nonlocality) in Continuous-Variable (CV) quantum systems.

• The Limitation of Standard CV Systems: It is well-established that Gaussian CV systems (e.g., optical fields with positive Wigner functions) cannot violate noncontextuality inequalities using standard quadrature measurements (homodyne detection). This is because such measurements can be described by a noncontextual hidden-variable model, and the destructive nature of quadrature measurements prevents the sequential measurements required to reveal contextuality.
• The Gap: While Discrete-Variable (DV) systems (qubits) have successfully demonstrated loophole-free contextuality tests, the CV counterpart remains largely unexplored due to the difficulty of performing non-destructive, sharp sequential measurements on continuous spectra without compromising measurement precision.
• The Goal: To experimentally demonstrate a violation of a Bell-like noncontextuality inequality in a CV system using a "black-box" approach that circumvents the limitations of standard quadrature detection.

2. Methodology

The authors propose and realize a hybrid encoding scheme combining discrete and continuous variables to perform a Hadamard test instead of direct sequential measurements.

A. Theoretical Framework: CV Peres-Mermin Square

• The experiment is based on the Peres-Mermin square, a state-independent proof of contextuality.
• The authors map the Pauli operators of the standard DV Peres-Mermin square to displacement operators ($D_x, D_y$) in the Gottesman-Kitaev-Preskill (GKP) code space.
• The inequality tested is:
  $$L = \left| -\sum_{k=1}^3 \langle O_{1k}O_{2k}O_{3k} \rangle + \sum_{j=1}^3 \langle O_{j1}O_{j2}O_{j3} \rangle \right| \leq 3\sqrt{3} \approx 5.196$$
  (Note: The paper cites the bound as $3\sqrt{3}$, but the experimental result $L \approx 5.94$ exceeds this, violating the noncontextual bound).
• Key Insight: Instead of measuring the displacement operators directly (which would destroy the state), the authors use the Hadamard test. This allows them to estimate the real part of the expectation value of the product of operators by using an ancilla qubit.

B. Hybrid Encoding System

• Qubit (Ancilla): Encoded in the polarization of a single photon ($|H\rangle \leftrightarrow |0\rangle$, $|V\rangle \leftrightarrow |1\rangle$).
• CV System: Encoded in the spatial modes (transverse position and momentum) of the same single photon. Under the plane-wave approximation, the 2D spatial wavefunction $\Psi(x, y)$ acts as a two-mode CV system.
• Source: Deterministic single photons are generated from an InAs/GaAs quantum dot embedded in a suspended membrane, excited by a pulsed laser. The source exhibits high purity ($g^{(2)}(0) \approx 0.0083$), ensuring no multiphoton events that would confuse the Hadamard test outcomes.

C. Experimental Implementation

• Controlled Displacements: The experiment requires controlled $q$- (position) and $p$- (momentum) displacements conditioned on the photon's polarization.
  • $q$-displacement: Realized using beam displacers (birefringent crystals) that shift the spatial path of $|H\rangle$ and $|V\rangle$ photons by a distance $q_0 = 3$ mm.
  • $p$-displacement: Realized using wedge plates that introduce a linear phase gradient, equivalent to a small angular deflection ($p_0 \approx 78.3 \, \mu\text{rad}$).
• Condition for Anti-commutativity: To mimic Pauli anti-commutation, the displacement parameters are calibrated such that $q_0 p_0 = \pi/2$.
• Measurement: The Hadamard test is performed by preparing the ancilla in a superposition, applying the sequence of controlled displacement operators, and measuring the ancilla polarization in the $|\pm\rangle$ basis. The probability of measuring $|-\rangle$ relates to the real part of the expectation value.

3. Key Contributions

1. First CV Contextuality Violation: This is the first experimental demonstration of a violation of a noncontextuality inequality in a CV system using a black-box-style approach.
2. Hadamard Test for CV Systems: The authors successfully adapted the Hadamard test to CV systems by mapping logical operations to spatial mode displacements of single photons. This bypasses the need for destructive quadrature measurements.
3. Hybrid Encoding: The work demonstrates a robust method for encoding CV information in the spatial degrees of freedom of deterministic single photons, controlled by a polarization qubit.
4. Commutativity Verification: The team implemented a "quantum switch"-like setup to verify that the realized displacement operators commute pairwise within the same context, a necessary condition for the validity of the inequality test.

4. Results

• Violation of Inequality: The experiment yielded a value of $L = 5.9398 \pm 0.0019$.
• Statistical Significance: This result violates the noncontextual hidden-variable bound ($3\sqrt{3} \approx 5.196$) by 380 standard deviations.
• Commutativity Check: The degree of non-commutativity ($\kappa$) between operator pairs was measured to be very low, with a mean value of $(1.74 \pm 0.30) \times 10^{-2}$ and a maximum instance below 0.023. This confirms that the experimental imperfections (non-commutativity) were not the cause of the violation.
• Source Purity: The single-photon source achieved a $g^{(2)}(0)$ of $0.83\%$, ensuring the binary outcomes required for the Hadamard test were well-defined.

5. Significance

• Foundational Physics: The results refute the idea that CV systems with positive Wigner functions are inherently classical in the context of sequential measurements. It demonstrates that quantum contextuality is a universal feature, accessible even in CV regimes when appropriate encoding and measurement strategies are used.
• Quantum Computing: The methodology supports the development of fault-tolerant quantum computing using GKP codes. By showing that GKP-like operations can be realized and tested on photonic platforms, it bridges the gap between theoretical error correction codes and physical implementation.
• New Experimental Paradigm: The "black-box" approach using Hadamard tests offers a pathway to test hidden-variable models in CV systems without the stringent requirements of non-Gaussian state preparation or post-selection, potentially paving the way for loophole-free tests in the future.
• Hybrid Systems: The work highlights the power of hybrid DV-CV systems, suggesting that using auxiliary modes (like spatial degrees of freedom) of deterministic photons is a viable and powerful resource for quantum information processing.

In summary, this paper successfully overcomes the historical barrier preventing the observation of contextuality in Gaussian CV systems by utilizing a hybrid photonic architecture and the Hadamard test, providing strong experimental evidence for the nonclassical nature of continuous-variable quantum mechanics.