Clifford-angle hardware-efficient ansatz (4 qubits), referencing arXiv:2607.15076
This card records a deterministic simulator run on Provenova inspired by this paper. It does not reproduce the paper's hardware results, and does not imply any endorsement by its authors.
Primitive: a four-qubit hardware-efficient variational ansatz — two layers of single-qubit ry/rz rotations with a linear cx entangling chain, plus a final rotation layer — in which every rotation angle is a Clifford angle (an integer multiple of pi/2).
What the referenced paper says. The abstract of arXiv:2607.15076 states that Subspace Quantum Diagonalization recovers ground-state energies by classically diagonalizing a Hamiltonian in the subspace spanned by quantum samples, requiring only bitstrings with sufficient ground-state overlap rather than an accurate variational energy. The authors say they exploit this robustness by asking how much non-Clifford and variational expressivity can be removed from the sampling circuit before accuracy degrades, and answer with two compression techniques: gradient-based operator pruning, which discards low-impact excitation operators, and Clifford rounding, which snaps remaining parameters to the nearest Clifford angle. The abstract states that both techniques can be applied to a VQE ansatz on a qubit-reduced Hamiltonian, reports an ablation study across 21 molecules, and reports hardware validation on 6 molecules on IBM quantum hardware. Those reported figures are the authors' and are deliberately not restated as numbers here.
What this card actually is. This card references that paper and is not an implementation of the authors' pipeline, not a chemistry calculation, and not endorsed by them. It illustrates only the structural idea named in the abstract — rotation parameters snapped to Clifford angles — on a generic textbook hardware-efficient ansatz with no molecule, no Hamiltonian, and no optimization. Every angle here is 0, pi/2, pi or -pi/2, so the circuit is entirely Clifford and the prepared state is a stabilizer state.
Provenance. The counts on this card come from a deterministic, seeded simulation run on Provenova (4096 shots, seed 1729). This is a simulator run, not a reproduction of the paper's hardware results or of its accuracy and speedup findings. The circuit hash and seed make the run independently repeatable.
Verify offline — this hash is Merkle-bound to the exact calibration and hardware state. How verification works →
[](https://provenova.net/cards/clifford-angle-hardware-efficient-ansatz-4-qubit-8nqqx1j0)
<iframe src="https://provenova.net/cards/clifford-angle-hardware-efficient-ansatz-4-qubit-8nqqx1j0/embed.html" width="400" height="420" style="border:0;overflow:hidden" loading="lazy" title="Clifford-angle hardware-efficient ansatz (4 qubits), referencing arXiv:2607.15076 — Provenova"></iframe>
The vendor-neutral system of record for quantum — every run bound to the exact calibration that produced it, reproducible and offline-verifiable.