Pith. sign in

REVIEW 4 major objections 6 minor 2 cited by

Mid-circuit measurement as an algorithmic primitive

T0 review · 4 major / 6 minor · reviewed 2026-08-07 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗

Pith's one-line read This paper's central claim is that a mid-circuit measurement on a single ancilla acts as a low-energy filter, amplifying the ground state of a diagonal cost Hamiltonian and steering the system toward optimal solutions without a classical…

desk verdict Correct but standard cosine filter, demonstrated with oracle knowledge of the answer; worth a referee for the repair path, not for the current claims. read the letter →

arxiv 2506.00118 v2 pith:KKSS62CI submitted 2025-05-30 quant-ph

Antoine Lemelin , Christophe Pere , Olivier Landon-Cardinal , Camille Coti This is my paper
classification quant-ph
keywords mid-circuitmeasurementquantumphaseestimationHadamardtestenergyfilterQAOAvertexcoveramplitudeamplificationfeedback
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

This paper tries to establish that a single mid-circuit measurement of an ancilla qubit is not only an error-correction tool but an algorithmic primitive: a low-energy filter that amplifies the ground state of a diagonal cost Hamiltonian. The authors show that when the Hadamard-test ancilla returns 0, each energy amplitude is multiplied by a cosine that is maximal at the lower energy bound and zero at the upper bound, and when it returns 1, a fixed mixer rotation heuristically restores overlap with low-energy states. They argue that this replaces the classical outer optimization loop of variational algorithms with a purely quantum feedback loop, and they report numerical exponential convergence of the probability of measuring optimal vertex covers, with a 4-qubit hardware run showing the same trend. If the claim is right, mid-circuit measurement becomes a way to steer near-term quantum optimization without variational parameter tuning.

What carries the argument

The load-bearing object is the cosine filter identity, the amplitude transformation $c'_k = c_k\cos(\vartheta_k)\big/\sqrt{\sum_j |c_j|^2\cos^2\vartheta_j}$ with $\vartheta_k = \frac{\pi}{2}\frac{E_k-E_{\rm inf}}{E_{\rm sup}-E_{\rm inf}}$. This is the exact projection produced by the single-ancilla phase estimation circuit when the ancilla outcome is 0. The paper sets the ancilla rotation $\theta = -\pi E_{\rm inf}/(E_{\rm sup}-E_{\rm inf})$ and the evolution time $t = -\pi/(E_{\rm sup}-E_{\rm inf})$ so that the cosine equals 1 at the lower bound and 0 at the upper bound. The companion identity for outcome 1 replaces cosine with sine, and the fixed $\pi/2$ x-rotation on every system qubit is the heuristic repair step that returns amplitude to the low-energy sector; repeated alternation of these two operations is the algorithm.

What would settle it

Run the iteration loop on a graph with the lower bound deliberately set below the true ground energy, for instance $E_{\rm inf}=E_0-\Delta$, and measure the probability of sampling the optimum as a function of iterations; if amplification is insensitive to $\Delta$, the method truly does not need the exact ground energy, whereas strong sensitivity would show that the demonstrated convergence relies on the NP-hard value used in the pseudocode.

Watch

Extended reading notes

Core claim

Starting from a state $|\psi\rangle=\sum_k c_k|E_k\rangle$ in the eigenbasis of a Hamiltonian diagonal in the computational basis, the authors apply the Hadamard-test circuit that controls $U=e^{-iHt}$ with an ancilla rotation $R_z(\theta)$. Measuring the ancilla in $|0\rangle$ projects the system into a state whose amplitudes are $c'_k\propto c_k\cos((\theta-tE_k)/2)$. They choose $\theta=-\pi E_{\rm inf}/(E_{\rm sup}-E_{\rm inf})$ and $t=-\pi/(E_{\rm sup}-E_{\rm inf})$ so that the cosine is 1 at the lower energy bound and 0 at the upper bound, and therefore the low-energy sector is amplified while the high-energy sector is suppressed. When the ancilla reads $|1\rangle$, the same calculation gives $c'_k\propto c_k\sin((\theta-tE_k)/2)$, which suppresses the low-energy sector; the paper proposes applying an $R_x(\pi/2)$ rotation to every qubit as a fixed mixer that heuristically restores overlap with low energies. The numerical evidence is that for random graphs with 3 to 11 qubits, the probability of measuring a minimum vertex cover grows exponentially in the number of iterations, and a run on real quantum hardware with 4 qubits shows the same trend without reaching full convergence.

Load-bearing premise

The load-bearing assumption is that a lower bound $E_{\rm inf}$ on the Hamiltonian spectrum is available before the search; in the pseudocode and numerics this bound is set to the exact ground energy $E_0$ computed from the known minimum vertex cover size $k_{\rm min}$, which is the NP-hard value the algorithm is supposed to find.

Editorial extensions

If this is right

  • A QAOA-style circuit can be driven toward the ground state without any classical optimizer, replacing the usual product of optimization iterations and shots per iteration with a single shot count.
  • For the tested random vertex-cover instances, the probability of sampling an optimal solution grows exponentially in the number of interference iterations, so the method offers a parameter-free alternative to variational training at moderate sizes.
  • Protecting only the ancilla qubit from noise substantially improves success probability, which points to targeted error mitigation on the measurement-feedback qubit as the scarce resource.
  • Because the Hamiltonian is required only to be diagonal with known energy bounds, any QUBO problem with such bounds inherits the same amplitude-shaping mechanism.
  • The authors explicitly note that the exponential iteration count does not bypass NP-hardness, so the value of the primitive lies in robust heuristics on near-term hardware rather than in complexity-theoretic speedups.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • A natural test is to run the algorithm with a cheap lower bound on the ground energy rather than the exact value; the convergence rate as the bound loosens would quantify how much information the filter actually needs.
  • The cosine filter can be viewed as measurement-based amplitude amplification with a continuous profile rather than Grover's sign flip; comparing its iteration count with standard amplitude-amplification scaling would locate the method in that landscape.
  • The protected-ancilla result predicts that on current hardware, error mitigation focused on the ancilla should recover most of the ideal success probability; this is directly testable before any full error-correction overhead is introduced.
  • Although the paper assumes a Hamiltonian diagonal in the computational basis, the interference argument should transfer to any Hamiltonian whose eigenbasis is reachable by a known unitary; applying the filter in that rotated basis would be a natural extension.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

4 major / 6 minor

Summary. The paper proposes a quantum optimization algorithm for diagonal (Ising/QUBO) Hamiltonians in which a single-ancilla Hadamard test is used as a low-energy filter: conditioning on the ancilla outcome 0 multiplies eigenstate amplitudes by a cosine that peaks at a lower energy bound, while the undesired outcome is followed by an Rx(π/2) mixer designed to repopulate low-energy states. The authors derive the post-selection amplitude transformation (Eqs. 5–9), give a QUBO formulation for minimum vertex cover, present numerical convergence curves up to n=11, and report a 7-iteration run on ibm_quebec. The central claim is that mid-circuit measurement can serve as an algorithmic primitive that amplifies ground-state population without a classical optimization loop and with no knowledge of the ground energy E0.

Significance. If the claims were fully substantiated, the paper would provide a simple, hardware-compatible primitive for measurement-based optimization, with credible potential to reduce reliance on classical variational loops. The mathematical derivation of the post-selection cosine filter is correct and clearly presented, and the explicit QUBO instance is concrete. However, the load-bearing numerical evidence is currently obtained using the exact ground energy E0 (derived from kmin, the NP-hard target) to set the filter parameters, and the mixer is a heuristic with no supporting analysis. The paper also lacks statistical detail for the convergence curves, and the hardware demonstration does not converge. The conceptual contribution is worth pursuing, but the evidence in this version does not establish the claimed regime.

major comments (4)
  1. [§2.3 / Pseudocode lines 4–5 / §2.4.6] The algorithm's stated operating regime (Hypothesis 2) only assumes known lower and upper bounds Einf and Esup, not the exact eigenvalues. Yet the pseudocode sets θ = −πE0/(Emax−E0) and t = −π/(Emax−E0), and §2.4.6 computes E0 = A(2kmin−n) using kmin, the size of a minimum vertex cover—the very quantity the algorithm is meant to find. This means the numerical convergence in Fig. 8 is obtained with oracle knowledge of the answer, so it cannot support the 'no knowledge of E0' claim. Please provide a polynomially computable Einf (or otherwise specify a legitimate bound) and rerun the numerical study without using kmin; alternatively, clearly restrict the claims to a setting in which E0 is known.
  2. [§2.2] The mixer U' = ⊗ Rx(π/2) · U is introduced with the heuristic statement that it 'maps the high-energy sector to the low-energy sector.' This statement is load-bearing: after an undesired measurement the state has increased high-energy amplitudes (Eq. 9), and the convergence observed in Fig. 8 depends on the mixer restoring low-energy overlap. No argument or quantitative evidence is given for this mapping for the vertex cover Hamiltonians used, and the text does not specify the class of Hamiltonians for which the heuristic holds. The paper should either prove a bound for a well-defined problem class or provide numerical experiments that isolate and quantify the mixer's effect, including cases where the mixer is omitted.
  3. [§3.2, Fig. 8] The central numerical claim of exponential convergence is presented without error bars, numbers of shots, standard deviations, or details of the random graph generation. The text itself notes that instances with unique optima have high variance, so the absence of statistical reporting makes it impossible to assess whether the observed curves are representative. Please report shot counts and repeated-seed statistics, and preferably fit the convergence rate with confidence intervals.
  4. [§3.4, Fig. 10] The abstract states the approach is validated on real quantum hardware, but the hardware execution over 7 iterations does not achieve convergence, as the text concedes. A non-converging run demonstrates circuit execution but does not validate the algorithm's amplifying behavior on hardware. Please reframe this section as a feasibility/execution study or provide additional data (e.g., comparison with simulation under a noise model, or longer runs) that supports the validation claim.
minor comments (6)
  1. [§2.3] Equations (12) and (13) are referred to as numbered equations but are not actually numbered in the text; please number them or adjust the cross-references.
  2. [Introduction / §2.1] There are several typos and garbled phrases, including 'where ϵ is 1 2' in the Introduction and 'aancilla' in §2.1.
  3. [Figure 2 caption] The caption uses 'Estimated Esup' while the text refers to Emax; please make the notation consistent.
  4. [§2.4] The sentence 'By shifting the energy by -1' is not explained, and the subsequent E0 formula gives E0=0 for the example, which is confusing.
  5. [§3.2, Eq. (24)] The phrase 'where |ψk⟩ is a Good Answer' is informal; please define 'Good Answer' explicitly and specify whether Pgood includes all optimal solutions.
  6. [Pseudocode] The pseudocode is referred to as 'Pseudocode 0' but has no caption or label; please add one and reference it consistently.

Circularity Check

1 steps flagged · score 6.0 of 10

Numerical demonstration is circular: cosine-filter parameters use the exact ground energy E0 obtained from kmin, so the observed convergence does not test the claimed 'bounds-only' regime.

  1. self definitional [Pseudocode lines 4–5; Sec. 2.4.6; contradicting Hypothesis 2 (Sec. 2.1.1)]
    "Please note that we do not assume to know precisely the energy minimum E0, nor the maximum energy Emax. ... Apply Rz(θ) on ancilla, with θ = −πE0 Emax−E0 ... Apply controlled-U = e−iHt with t = −π Emax−E0 ... The minimum energy E0 occurs for a valid minimum vertex cover Cmin of size kmin, with xi = +1 for i ∈ Cmin and xi = −1 otherwise: E0 = A(2kmin − n)."

    Hypothesis 2 defines the method's required input as a lower bound Einf, explicitly not the exact ground energy E0. But the pseudocode sets θ and t using E0 and Emax, and Sec. 2.4.6 computes E0 from kmin, the size of a minimum vertex cover—the very quantity the algorithm is supposed to find. The cosine filter is therefore centered on the answer by construction, so the convergence reported in Fig. 8 does not demonstrate the advertised 'no knowledge of E0' regime. The paper itself admits that the tightness of Einf impacts convergence, yet the simulations substitute exact E0 for Einf; the observed amplification is manufactured by oracle input rather than predicted from the stated assumptions.

full rationale

The core filter identity (Eqs. 4–8) is an algebraic consequence of the Hadamard-test post-selection and is not circular: after measuring |0⟩, amplitudes are multiplied by cos(π/2 · (Ek − Einf)/(Esup − Einf)), which follows from the circuit and the stated bound inputs. No self-citation is load-bearing; the references are standard external works. The circularity is confined to the implementation and benchmark: the pseudocode and Sec. 2.4.6 replace the promised lower bound Einf with the exact ground energy E0, itself obtained from kmin, the NP-hard target value. Consequently, the numerical evidence of exponential convergence to the optimal vertex cover is forced by a parameter choice that already encodes the answer. This does not invalidate the mathematical filter, but it means the paper's central demonstration does not support the claimed regime in which only a polynomially computable bound, and not E0, is known. A moderate circularity score of 6 reflects that one key 'prediction' reduces by construction to its input, while the underlying derivation retains independent content.

Assumptions & free parameters 4 free parameters · 4 assumptions · 0 invented entities

The central derivation relies on the standard cosine/sine postselection of a Hadamard test, plus two domain assumptions about spectral bounds. The only genuinely ad hoc element is the mixer heuristic. The free parameters are the QUBO weights, the mixer angle, and especially the exact ground energy used to set the filter constants.

free parameters (4)
  • kmin (minimum vertex cover size) = 2 for the 4-node graph, varied per random instance
    Used in Section 2.4.6 to compute the exact ground energy E0 = A(2 kmin - n), which then sets theta and t in the pseudocode. This is the optimal solution, not a generic bound.
  • Energy bounds Einf and Esup = Set to E0 and Emax in the example and pseudocode
    The theory requires known bounds; the implementation takes them to be the exact extremal energies, which requires solving the problem. A generic lower bound is not provided.
  • QUBO weights A and B = A=1, B=2
    Chosen in Section 2.4.3 to satisfy B > A/2 and keep integer coefficients. Standard but still hand-picked.
  • Mixer angle sigma = pi/2
    Fixed heuristically in Eq. (11); the paper says the angle may vary with the Hamiltonian, so it is an unoptimized free choice.
assumptions (4)
  • domain assumption Hypothesis 1: the eigenbasis of H is the computational basis.
    Stated in Section 2.1. It holds for QUBO/Ising Hamiltonians but restricts the method to diagonal cost Hamiltonians.
  • domain assumption Hypothesis 2: known lower and upper bounds Einf and Esup on the spectrum.
    Stated in Section 2.1. The practical version replaces these bounds with exact E0 and Emax, which is a much stronger assumption.
  • ad hoc to paper The Rx(pi/2) mixer maps high-energy states to low-energy states after an undesired measurement.
    Introduced heuristically in Section 2.2 with no proof; Figure 7 shows only a single instance.
  • standard math Ideal projective measurement and perfect controlled-e^{-iHt} evolution.
    Assumed in the derivation of Eqs. (3)-(9), standard in quantum algorithm analysis.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Mid-circuit measurement as an algorithmic primitive." pith.science (2026). https://pith.science/paper/KKSS62CI

@misc{pith2026250600118,
  author       = {Pith},
  title        = {Pith review of: Mid-circuit measurement as an algorithmic primitive},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/KKSS62CI}},
  note         = {Machine review of arXiv:2506.00118}
}
read the original abstract

We explore the usefulness of mid-circuit measurements to enhance quantum algorithmics. Specifically, we assess how quantum phase estimation (QPE) and mid-circuit measurements can improve the performance of variational quantum algorithms. Our focus is on the single-qubit version of QPE namely, the Hadamard test applied to the Quantum Approximate Optimization Algorithm (QAOA) ansatz. We demonstrate that a mid-circuit measurement acts as a low-energy filter when the desired outcome is obtained. When the other outcome is measured we heuristically rely on the mixer to repopulate the low energy states. Numerical simulations show that this method effectively amplifies the ground state. We validate our approach on real quantum hardware namely the IBM Quantum system one ibm_quebec.

Discussion (0). Sign in to comment.

Forward citations

Cited by 2 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

  1. Rapid Cavity-Based Mid-Circuit Measurement and Feedforward in a Neutral Atom Array

    quant-ph 2026-06 conditional novelty 6.0 of 10

    Cavity-based technique achieves sub-100 μs mid-circuit measurement and feedforward in neutral atom arrays with sub-percent infidelity and minimal crosstalk.

  2. Principles of Quantum Optimization for Constrained Problems

    quant-ph 2026-07 conditional novelty 5.0 of 10

    Computational slowdown in constrained quantum optimization is attributed to the speed of entanglement restructuring, and the paper shows how constraints create (or avoid) the narrow spectral gaps where this restructur...

Reference graph

Works this paper leans on

19 extracted references · 7 canonical work pages · cited by 2 Pith papers

  1. [1]

    arXiv:1411.4028 [quant-ph]

    Farhi, E., Goldstone, J., Gutmann, S.: A quantum approximate optimization algorithm (arXiv:1411.4028) (2014) https://doi.org/10.48550/arXiv.1411.4028 . arXiv:1411.4028 [quant-ph]

    arXiv doi:10.48550/arxiv.1411.4028 2014

  2. [2]

    Nature Communications 5(1), 4213 (2014) https://doi.org/10.1038/ ncomms5213

    Peruzzo, A., McClean, J., Shadbolt, P., Yung, M.-H., Zhou, X.-Q., Love, P.J., Aspuru-Guzik, A., O’Brien, J.L.: A variational eigenvalue solver on a quantum processor. Nature Communications 5(1), 4213 (2014) https://doi.org/10.1038/ ncomms5213 . arXiv:1304.3061 [quant-ph]

    arXiv 2014

  3. [3]

    Nature Reviews Physics 7(4), 174–189 (2025) https://doi

    Larocca, M., Thanasilp, S., Wang, S., Sharma, K., Biamonte, J., Coles, P.J., Cincio, L., McClean, J.R., Holmes, Z., Cerezo, M.: Barren plateaus in variational quantum computing. Nature Reviews Physics 7(4), 174–189 (2025) https://doi. org/10.1038/s42254-025-00813-9 . arXiv:2405.00781 [quant-ph]

    arXiv 2025

  4. [4]

    Nature Communications 9(1), 4812 (2018) https://doi.org/10.1038/s41467-018-07090-4

    McClean, J.R., Boixo, S., Smelyanskiy, V.N., Babbush, R., Neven, H.: Barren plateaus in quantum neural network training landscapes. Nature Communications 9(1), 4812 (2018) https://doi.org/10.1038/s41467-018-07090-4

    doi:10.1038/s41467-018-07090-4 2018

  5. [5]

    Russian Mathematical Surveys 52(6), 1191 (1997) https://doi.org/10.1070/ RM1997v052n06ABEH002155

    Kitaev, A.Y.: Quantum computations: algorithms and error correction. Russian Mathematical Surveys 52(6), 1191 (1997) https://doi.org/10.1070/ RM1997v052n06ABEH002155

    work page 1997

  6. [6]

    Physical Review A 52(4), 2493–2496 (1995) https://doi.org/10.1103/PhysRevA

    Shor, P.W.: Scheme for reducing decoherence in quantum computer memory. Physical Review A 52(4), 2493–2496 (1995) https://doi.org/10.1103/PhysRevA. 52.R2493

    doi:10.1103/physreva 1995

  7. [7]

    Semiclassical Shor's Algorithm

    Giorda, P., Iorio, A., Sen, S., Sen, S.: Semiclassical shor’s algorithm. Physical Review A 70(3), 032303 (2004) https://doi.org/10.1103/PhysRevA.70.032303 . arXiv:quant-ph/0303037

    work page Pith review arXiv 2004

  8. [8]

    Semiclassical Fourier Transform for Quantum Computation

    Griffiths, R.B., Niu, C.-S.: Semiclassical fourier transform for quantum computa- tion. Physical Review Letters 76(17), 3228–3231 (1996) https://doi.org/10.1103/ PhysRevLett.76.3228 . arXiv:quant-ph/9511007

    work page Pith review arXiv 1996

Show all 19 references
  1. [9]

    SIAM Journal on Computing 26(5), 1484–1509 (1997) https://doi.org/10.1137/S0097539795293172

    Shor, P.W.: Polynomial-time algorithms for prime factorization and discrete log- arithms on a quantum computer. SIAM Journal on Computing 26(5), 1484–1509 (1997) https://doi.org/10.1137/S0097539795293172 . arXiv:quant-ph/9508027

    1997 arXiv

  2. [10]

    Physical Review Letters 127(10), 100501 (2021) https://doi.org/10.1103/PhysRevLett.127.100501

    Corcoles, A.D., Takita, M., Inoue, K., Lekuch, S., Minev, Z.K., Chow, J.M., Gambetta, J.M.: Exploiting dynamic quantum circuits in a quantum algorithm with superconducting qubits. Physical Review Letters 127(10), 100501 (2021) https://doi.org/10.1103/PhysRevLett.127.100501 . a...

    2021 arXiv

  3. [11]

    arXiv:2403.09514 [quant-ph]

    B¨ aumer, E., Tripathi, V., Seif, A., Lidar, D., Wang, D.S.: Quantum fourier trans- form using dynamic circuits (arXiv:2403.09514) (2024) https://doi.org/10.48550/ arXiv.2403.09514 . arXiv:2403.09514 [quant-ph]

    2024 doi arXiv

  4. [12]

    arXiv:quant-ph/9511026

    Kitaev, A.Y.: Quantum measurements and the abelian stabilizer prob- lem (arXiv:quant-ph/9511026) (1995) https://doi.org/10.48550/arXiv.quant-ph/ 9511026 . arXiv:quant-ph/9511026

    1995 doi arXiv

  5. [13]

    Proceedings of the Royal Society of London

    Cleve, R., Ekert, A., Macchiavello, C., Mosca, M.: Quantum algorithms revisited. Proceedings of the Royal Society of London. Series A: Mathematical, Physical and Engineering Sciences 454(1969), 339–354 (1998) https://doi.org/10.1098/ rspa.1998.0164 . arXiv:quant-ph/9708016

    1969

  6. [14]

    arXiv:2505.15913 [quant-ph]

    Faehrmann, P.K., Eisert, J., Kueng, R.: In the shadow of the hadamard test: Using the garbage state for good and further modifications (arXiv:2505.15913) (2025) https://doi.org/10.48550/arXiv.2505.15913 . arXiv:2505.15913 [quant-ph]

    2025 doi

  7. [15]

    Quantum 5, 566 (2021) https://doi.org/10.22331/q-2021-10-19-566

    Rall, P.: Faster coherent quantum algorithms for phase, energy, and amplitude estimation. Quantum 5, 566 (2021) https://doi.org/10.22331/q-2021-10-19-566 . arXiv:2103.09717 [quant-ph]

    2021 arXiv

  8. [16]

    Nature 471(7336), 87–90 (2011) https://doi.org/10.1038/ nature09770

    Temme, K., Osborne, T.J., Vollbrecht, K.G., Poulin, D., Verstraete, F.: Quantum metropolis sampling. Nature 471(7336), 87–90 (2011) https://doi.org/10.1038/ nature09770 . arXiv:0911.3635 [quant-ph]

    2011 arXiv

  9. [17]

    Frontiers in Physics 2 (2014) https://doi.org/10.3389/fphy.2014.00005

    Lucas, A.: Ising formulations of many np problems. Frontiers in Physics 2 (2014) https://doi.org/10.3389/fphy.2014.00005

    2014

  10. [18]

    arXiv:quant-ph/9605043

    Grover, L.K.: A fast quantum mechanical algorithm for database search (arXiv:quant-ph/9605043) (1996) https://doi.org/10.48550/arXiv.quant-ph/ 9605043 . arXiv:quant-ph/9605043

    1996 doi arXiv

  11. [19]

    https://docs.quantum.ibm.com/ api/qiskit-ibm-runtime/docs.quantum.ibm.com/api/qiskit-ibm-runtime/ fake-provider-fake-quebec 23

    IBMQuantum: FakeQuebec (v1.4). https://docs.quantum.ibm.com/ api/qiskit-ibm-runtime/docs.quantum.ibm.com/api/qiskit-ibm-runtime/ fake-provider-fake-quebec 23

Pith tools

Reviewed August 7, 2026 · model on record in the stance chip above.