A new theoretical result, not yet peer reviewed, gives physicists an exact way to measure 'quantum magic' — a quantum state's distance from the easy to simulate classical regime — across phase transitions, extending a benchmarking framework that previously worked only for…
Physicists trying to estimate how hard a quantum system is to simulate on a classical computer have long leaned on entanglement entropy, a single number that grows predictably as a quantum phase transition approaches. A new theoretical analysis, posted to arXiv this month by Reyhaneh Khasseh, E. A. Ramirez Trino, and M. A. Rajabpour, extends that benchmarking framework to a complementary measure called stabilizer entropy, the entropy of the most classical-like quantum states a system can occupy. The result gives researchers an exact finite-size formula for this resource across the full critical-to-noncritical crossover, not only at the phase transition itself.[^1]
The probe in question is the stabilizer Rényi entropy at index α = 1/2, often shorthanded as quantum magic or nonstabilizerness. Quantum magic captures the part of a quantum state's behavior that cannot be reproduced by Clifford circuits, the relatively easy-to-simulate class of quantum operations that define the boundary of what classical computers can handle.[^2] Tracking how magic grows, or fails to grow, across a phase transition tells physicists and quantum engineers something entanglement entropy alone cannot: how far a given state sits from the classically tractable regime, and what it would cost a conventional computer to reproduce.
Until now, exact results for stabilizer entropy existed mainly at critical points, the precise tuning of a Hamiltonian where a system undergoes a continuous phase transition. There, conformal field theory supplies scaling laws. Away from the critical point, in the massive or noncritical regime, the scaling is messier and harder to pin down analytically. The new paper closes that gap for a broad class of finite-range spin chains, including transverse-field Ising, XX, free-fermion, and Potts-type models.[^3]
The central technical result is a clean decomposition of the stabilizer entropy in two geometries: a full periodic chain, a ring of spins where the system wraps around on itself, and a finite interval embedded in an infinite chain. In the periodic case, the entropy obeys a volume law away from criticality, meaning it grows proportionally to the number of spins in the system rather than just its boundary.[^4] A universal crossover structure sits at the critical-to-noncritical transition, and across that crossover subleading terms carry fingerprints of the correlation length—the characteristic distance over which disturbances in the system spread—and the boundary geometry.[^5] Those are the same kinds of fingerprints that have made entanglement entropy such a useful probe of quantum matter.
The practical payoff is a benchmark a physicist or simulation engineer can read off a single finite-size simulation. Where previously one had asymptotic critical-point formulas, this analysis yields exact numbers for the volume-law coefficient, the boundary-constant contribution, and the crossover width across the entire phase diagram.[^6] That makes stabilizer entropy a more honest quantity to measure when estimating classical-simulation cost or diagnosing whether a noisy intermediate-scale quantum (NISQ) device is producing states that strain classical simulators.
The authors are careful about scope. The universal formulas are exact for the specific models treated: the anisotropic XY chain and its transverse-field Ising and XX limits, zero-field Ising-block reductions, folded finite-range string Hamiltonians, and Cluster–Ising representatives with three-spin cluster interactions.[^7] They do not automatically transfer to long-range interactions, quenched disorder, or open-system settings where dissipation and decoherence are present. The work is best read as a benchmark for an important class of clean, closed quantum systems, not a closed theory of magic in general.
The paper has been posted to arXiv under identifier 2606.13810 but has not yet been peer-reviewed, and the authors have not registered a journal-version DOI.[^8] The result is best understood as a theoretical contribution awaiting community scrutiny. Watch for a journal submission in the coming months, replication or extension to long-range and disordered models, and whether the exact finite-size picture the authors establish holds up in numerical studies on larger systems.
[^1]: Reyhaneh Khasseh, E. A. Ramirez Trino, M. A. Rajabpour, "Universal Crossovers of Stabilizer Entropy Beyond Criticality," arXiv:2606.13810 (2026), https://arxiv.org/abs/2606.13810.
[^2]: bid.
[^3]: bid.
[^4]: bid.
[^5]: bid.
[^6]: bid.
[^7]: bid.
[^8]: bid; submission history shows only v1 from 11 June 2026 with no subsequent revisions or journal registration.