PERFORMANCE Signal 510
Spin audit of SQD/QSCI quantum-chemistry benchmarks on iron–sulfur clusters
↗
Illustration only Photo by Igor Omilaev on Unsplash
A spin-audit of SQD/QSCI quantum-chemistry benchmarks reveals published iron, sulfur cluster results did not converge to the claimed spin states.
Quantum chemistry simulations on near-term hardware rely on these benchmarks as proof of utility. If the spin state is wrong, the energy value is meaningless for chemical accuracy. Teams building or validating quantum algorithms must now add spin-moment checks to their benchmarking pipelines.
Written by elseif from the cluster below · every claim links back to a sourceThe three things worth knowing
Published SQD/QSCI iron, sulfur benchmarks converged to incorrect spin states at competitive energies.
The audit provides open data, code, and a reproduction map so any reader can verify the findings.
Spin moments are recommended as mandatory benchmark outputs for future quantum-chemistry demonstrations.
THE READ
What the cluster adds up to.
The event is a reproducibility audit of sample-based quantum diagonalization (SQD, also called QSCI) benchmarks on iron, sulfur clusters. These benchmarks were presented as evidence of utility-scale quantum chemistry, but the audit finds that none of the published runs converged to the claimed singlet spin state. Instead, the lowest-energy root at the largest published dimension has a spin expectation value far from a singlet. This discrepancy undermines the chemical relevance of the results, even if the energy values appeared competitive.
The audit introduces a rotation-invariant spin check and higher spin moments to verify convergence. These tools are not part of the original benchmarking workflow, meaning prior results were accepted without this validation. The audit’s open dataset and code allow independent verification, shifting the burden of proof to future benchmarking efforts. For engineers, this means adding spin-moment calculations to their quantum chemistry pipelines, which increases computational overhead but is necessary for chemical accuracy.
The audit’s findings show that hardware samples used in the original benchmarks offered no energy advantage over uniform-random controls. This suggests that the sampling method may not be capturing the intended electronic configurations. For teams building quantum algorithms, this implies that sample-based methods require additional validation beyond energy convergence. The audit’s recommendation to include spin moments as a standard benchmark output is a direct response to this gap.
The material provided is limited to a single source, so corroboration across feeds is absent. However, the audit’s open data and reproduction map are designed to enable independent verification, which is a stronger form of evidence than multiple headlines. The lack of additional coverage may reflect the niche nature of quantum chemistry benchmarks, but the audit’s implications are broad for anyone relying on these results to assess quantum hardware or algorithms.
Written by elseif from the cluster below · checked for specifics the sources never containedTHE CLUSTER