REVIEW
Nonlocal magic of fermions is bounded by entanglement fluctuations and is logarithmic for free fermions
Quantum Physics · Submitted 4 Nov 2026 · 66 references
Pith's verdict: Solid paper that proves nonlocal magic of free-fermion states is logarithmic, not linear, when parity-respecting non-Gaussian local unitaries are allowed, and gives a universal bound via capacity of entanglement.
Why Pith says this
Highlight any text to comment on it.
Referee report
The paper
The paper claims
Nonlocal magic of any pure fermionic state is bounded by the capacity of entanglement, and for free fermions it drops from linear to logarithmic.
The reading
What carries the argument
The capacity of entanglement $C_E = \operatorname{Var} H_E$ (variance of the entanglement Hamiltonian) is the central object. The paper derives an exact kernel identity linking the best flat-spectrum fidelity to the distribution of the entanglement energy $X = -\log_2\lambda$, and uses statistical inequalities (Berry–Esseen, concentration) to bound the fidelity by $C_E$. For free fermions, the additive structure of independent BCS pairs guarantees $C_E \le c_* m$, making the fidelity decay only polynomial in $m$.
Extended reading notes
Core claim
The central discovery is a universal bound on the nonlocal stabilizer fidelity: for every pure fermionic state of definite parity, $\Phi_{\mathrm{NLS}} \ge 0.478/\sqrt{C_E+5}$, where $C_E$ is the capacity of entanglement. Consequently, the nonlocal stabilizer Rényi entropy satisfies $M_{f,\mathrm{NL}}^2 \le 4 D_{f,\mathrm{NL}}^{\min} \le 2\log_2(C_E+5)+\text{constant}$. For free-fermion states, $C_E \le c_* m$ with $c_*\approx 0.44$, so $M_{f,\mathrm{NL}}^2$ is $O(\log m)$, whereas the Gaussian-restricted value $M_2(p)$ is $\Theta(m)$. The paper further establishes that the fidelity of $m$ BCS pairs to fermionic Bell pairs under optimal parity-respecting local unitaries decays as $1/\sqrt{m}$, in contrast to the exponential decay under Gaussian unitaries.
Load-bearing premise
The claim that free-fermion nonlocal magic is logarithmic relies on the capacity of entanglement growing at most linearly with the number $m$ of BCS pairs, which in turn follows from the independence of the pairs' entanglement energies; if correlations violated this independence, $C_E$ could grow faster and the bound would be weaker.
Editorial extensions
If this is right
- Almost all the magic of free-fermion states (including Hartree–Fock and BCS mean-field states) is local; only a logarithmic part is nonlocal and must be stored across a bipartition.
- If local parity-respecting unitaries are available, the number of $T$ gates needed to prepare a free-fermion state beyond local processing drops from $\Theta(m)$ to $O(\log m)$.
- For critical fermionic chains such as the XX model, the nonlocal magic grows only as $\log \log \ell$ with block size $\ell$, accessible only at very large $\ell$.
- The capacity bound provides an experimentally accessible way to bound nonlocal magic: measure $C_E$ via entanglement Hamiltonian tomography and apply the inequality $\Phi_{\mathrm{NLS}} \ge 0.478/\sqrt{C_E+5}$.
Reading between the lines
- The capacity bound likely extends to other quantum resource measures such as the nonlocal negativity or the entangled resource needed for state merging, offering a unifying connection between entanglement fluctuations and nonlocal resources.
- In holographic settings, where Newton's constant sets $C_E \sim 1/G_N$, the logarithmic bound ties nonlocal magic to $O(\log(1/G_N))$ bits, consistent with holographic expectations.
- Under strict particle-number conservation (no phase reference), the paper notes that fidelity-based nonlocal magic remains extensive for identical pairs; this suggests number-conserving local operations are significantly weaker at removing magic.
- The optimality of Gaussian unitaries for up to three pairs but not beyond provides a sharp crossover boundary that could be tested in small fermionic quantum processors.
Original abstract
Free-fermion states are classically simulable, yet, written as qubits, they typically carry extensive magic, the quantum resource that stabilizer circuits lack. Part of this magic is nonlocal: local unitaries acting separately on two parts of the system cannot remove it. If these local unitaries are restricted to Gaussian ones, the nonlocal magic grows linearly with the number $m$ of partially entangled fermion pairs. We show that this growth is an artifact of the Gaussian restriction: local Gaussian unitaries minimize the nonlocal stabilizer R\'enyi entropy for up to three pairs, but not from four pairs on. General local non-Gaussian unitaries that respect fermion parity take $m$ free-fermion pairs to fermionic Bell pairs with a fidelity that decays at most as $1/\sqrt m$ compared to exponential decay with Gaussian unitaries, so the nonlocal magic drops from linear to logarithmic in $m$. The reason is statistical --- Bell pairs have equal Schmidt weights, that is, a fully degenerate entanglement spectrum, and a state can stay far from them only if its Schmidt weights are spread over many orders of magnitude. This spread is measured by the capacity of entanglement $C_E$, the variance of the entanglement Hamiltonian, and the nonlocal magic of any fermionic state grows at most logarithmically with $C_E$. Free-fermion pairs add to $C_E$ independently, so $C_E$ grows only linearly in $m$, and this leads to the stabilizer fidelity decreasing at most as $1/\sqrt m$. Almost all the magic of free fermions, including Hartree--Fock and BCS mean-field states, is therefore local.
The compiled record
The paper's own sentences, arranged, kept beside the review and separate from it. Each label is drawn from a closed vocabulary; the number beside it is the probability the classifier gave that label. Numbers in the tables are the authors', checked by code. No sentence in this record was written by a model.
The abstract, sentence by sentence
Free-fermion states are classically simulable, yet, written as qubits, they typically carry extensive magic, the quantum resource that stabilizer circuits lack. Part of this magic is nonlocal: local unitaries acting separately on two parts of the system cannot remove it. If these local unitaries are restricted to Gaussian ones, the nonlocal magic grows linearly with the number $m$ of partially entangled fermion pairs. We show that this growth is an artifact of the Gaussian restriction: local Gaussian unitaries minimize the nonlocal stabilizer R\'enyi entropy for up to three pairs, but not from four pairs on. General local non-Gaussian unitaries that respect fermion parity take $m$ free-fermion pairs to fermionic Bell pairs with a fidelity that decays at most as $1/\sqrt m$ compared to exponential decay with Gaussian unitaries, so the nonlocal magic drops from linear to logarithmic in $m$. The reason is statistical --- Bell pairs have equal Schmidt weights, that is, a fully degenerate entanglement spectrum, and a state can stay far from them only if its Schmidt weights are spread over many orders of magnitude. This spread is measured by the capacity of entanglement $C_E$, the variance of the entanglement Hamiltonian, and the nonlocal magic of any fermionic state grows at most logarithmically with $C_E$. Free-fermion pairs add to $C_E$ independently, so $C_E$ grows only linearly in $m$, and this leads to the stabilizer fidelity decreasing at most as $1/\sqrt m$. Almost all the magic of free fermions, including Hartree--Fock and BCS mean-field states, is therefore local.
background contribution method result interpretation main claim
| # | sentence | role | the authors' own work | states the main conclusion | strength |
|---|---|---|---|---|---|
| s0 | Free-fermion states are classically simulable, yet, written as qubits, they typically carry extensive magic, the quantum resource that stabilizer circuits lack. | background 0.99 | 0.09 | 0.04 | asserted 0.99 |
| s1 | Part of this magic is nonlocal: local unitaries acting separately on two parts of the system cannot remove it. | background 0.79 | 0.17 | 0.03 | asserted 0.99 |
| s2 | If these local unitaries are restricted to Gaussian ones, the nonlocal magic grows linearly with the number $m$ of partially entangled fermion pairs. | background 0.68 | 0.32 | 0.05 | asserted 1.00 |
| s3 | We show that this growth is an artifact of the Gaussian restriction: local Gaussian unitaries minimize the nonlocal stabilizer R\'enyi entropy for up to three pairs, but not from four pairs on. | result 0.97 | 0.91 | 0.14 | asserted 1.00 |
| s4 | General local non-Gaussian unitaries that respect fermion parity take $m$ free-fermion pairs to fermionic Bell pairs with a fidelity that decays at most as $1/\sqrt m$ compared to exponential decay with Gaussian unitaries, so the nonlocal magic drops from linear to logarithmic in $m$. | result 0.98 | 0.90 | 0.69 | asserted 0.99 |
| s5 | The reason is statistical --- Bell pairs have equal Schmidt weights, that is, a fully degenerate entanglement spectrum, and a state can stay far from them only if its Schmidt weights are spread over many orders of magnitude. | interpretation 0.99 | 0.66 | 0.07 | asserted 1.00 |
| s6 | This spread is measured by the capacity of entanglement $C_E$, the variance of the entanglement Hamiltonian, and the nonlocal magic of any fermionic state grows at most logarithmically with $C_E$. | result 0.82 | 0.83 | 0.50 | asserted 1.00 |
| s7 | Free-fermion pairs add to $C_E$ independently, so $C_E$ grows only linearly in $m$, and this leads to the stabilizer fidelity decreasing at most as $1/\sqrt m$. | result 0.60 | 0.83 | 0.12 | asserted 1.00 |
| s8 | Almost all the magic of free fermions, including Hartree--Fock and BCS mean-field states, is therefore local. | interpretation 0.81 | 0.80 | 0.40 | asserted 0.97 |
What was examined, and what was not
- arXiv listing abstract v1
- full text
- figures and tables
- references and citing sentences
- relations to other papers
- P abstract
- S sections
- T body sentences
- C citations
- R cross-paper relations
Figures
Showing 1 of the paper's 3 figures. See all figures in the PDF ↗ Figures reproduced from arXiv:2611.07003 by the authors.
References
Discussion
Disagree with the review, or know something it missed? Leave a comment, highlight any passage above to comment on it, or challenge this review formally.