Computable fermionic non-Gaussianity from the covariance matrix

Barbara Kraus; Bernhard Jobst; Frank Pollmann; Marc Langer; Poetri Sonya Tarabunga; Ra\'ul Morral-Yepes; Sheng-Hsuan Lin

REVIEW 2 major objections 2 minor 2 cited by

A Tsallis entropy of fermionic occupation numbers is monotonic under Gaussian protocols and lower-bounds the number of non-Gaussian gates needed for state preparation.

Reviewed by Pith at T0; open to challenge. T0 means a machine referee read the full paper against a public rubric. the ladder, T0–T4 →

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

T0 review · grok-4.3

2026-07-03 12:17 UTC pith:IRWZIKWR

load-bearing objection The paper gives two new families of covariance-based measures for fermionic non-Gaussianity on pure states, with a monotonicity proof under Gaussian protocols for one Tsallis case. the 2 major comments →

arxiv 2607.02242 v2 pith:IRWZIKWR submitted 2026-07-02 quant-ph cond-mat.str-el

Computable fermionic non-Gaussianity from the covariance matrix

Poetri Sonya Tarabunga , Bernhard Jobst , Ra\'ul Morral-Yepes , Marc Langer , Barbara Kraus , Frank Pollmann , Sheng-Hsuan Lin This is my paper

classification quant-ph cond-mat.str-el

keywords fermionic non-Gaussianitycovariance matrixresource theoryTsallis entropyoccupation numbersnatural orbitalsGaussian operationsmany-body systems

verification ladder T0 review T1 audit T2 compute T3 formal T4 reserved

The pith

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

The paper develops a convex resource theory for fermionic non-Gaussianity and defines two families of measures for pure states that are computed from the covariance matrix. Occupation number entropies are Tsallis-α entropies of the occupation numbers obtained via the Williamson normal form; one specific member of this family is proven monotonic under Gaussian operations and therefore functions as a resource monotone. This monotone directly lower-bounds the minimal number of non-Gaussian gates required to prepare the state. Natural-orbital participation entropies are Rényi-α entropies of the squared amplitudes in the natural-orbital basis and upper-bound the cost of classical simulation in a Gaussian orthonormal basis. The measures are evaluated on stabilizer states, translation-invariant states, random circuits, and the XXZ model to illustrate their behavior in concrete settings.

Core claim

From the Williamson normal form of the covariance matrix of a pure fermionic state, occupation number entropies are defined as the Tsallis-α entropy of the resulting occupation numbers; one member of this family is monotonic under Gaussian protocols and hence a computable convex resource monotone for fermionic non-Gaussianity that lower-bounds the number of non-Gaussian gates needed for state preparation. A second family of natural-orbital participation entropies, given by the Rényi-α entropy of the squared amplitudes in the natural-orbital basis, quantifies compressibility and upper-bounds classical simulation cost.

What carries the argument

The occupation number entropies, obtained as the Tsallis-α entropy of the occupation numbers that appear in the Williamson normal form of the covariance matrix.

Load-bearing premise

The measures are defined only for pure fermionic states and assume that the covariance matrix alone encodes all information needed to quantify the relevant non-Gaussianity.

What would settle it

A concrete Gaussian operation that strictly increases the value of the proposed Tsallis monotone on some pure fermionic state, or an explicit state preparation sequence whose non-Gaussian gate count falls below the lower bound supplied by the monotone.

Watch this falsifier — get emailed when new claim-graph text bears on it.

If this is right

The monotone lower-bounds the number of non-Gaussian gates required to prepare any given pure fermionic state.
The natural-orbital measures upper-bound the classical simulation cost of the state in an orthonormal Gaussian basis.
Both families become simple explicit functions for stabilizer states and for translation-invariant states.
The measures can be evaluated on states generated by random SWAP-doped matchgate circuits and on ground states of the bond-modulated XXZ model.
The resource-theoretic framework supplies practical tools for assessing classical simulability of fermionic states.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

If the covariance matrix ceases to be sufficient for mixed states, the monotonicity and bounding properties would require additional correlation terms.
The same construction might be tested for its ability to detect non-Gaussianity in open-system dynamics where Gaussian noise is added.
The upper bound on simulation cost could be compared directly with existing tensor-network contraction costs for the same states.

Editorial analysis

A structured set of objections, weighed in public.

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

Desk Editor's Note

The paper gives two new families of covariance-based measures for fermionic non-Gaussianity on pure states, with a monotonicity proof under Gaussian protocols for one Tsallis case.

read the letter

The main contribution is the construction of two explicit families of measures pulled from the Williamson eigenvalues of the covariance matrix. The occupation-number family uses Tsallis-α entropy on those eigenvalues; they show one specific α yields a convex monotone under Gaussian operations and therefore lower-bounds the number of non-Gaussian gates. The second family uses Rényi-α entropy on the squared natural-orbital amplitudes and supplies an upper bound on classical simulation cost in that basis. Both are computable once the covariance is known, which is the practical point.

The paper checks the measures on stabilizer states, translation-invariant cases, random SWAP-doped matchgate circuits, and the bond-modulated XXZ model. Those calculations are straightforward and show the quantities behave as expected when non-Gaussianity is added or removed. The resource-theoretic framing is also laid out cleanly, with the monotonicity claim tied directly to the covariance.

The soft spot is the reliance on the covariance alone. For pure fermionic states the one-body reduced density matrix is fixed by the covariance, but higher-order correlators are not. If a Gaussian protocol can connect states that share the same covariance yet differ in those correlators, the measure stays fixed while the underlying non-Gaussian resource may change. The abstract indicates the proofs treat the covariance as sufficient, so the monotonicity and bounding statements rest on that step. Whether the proofs close the gap or simply define the resource via the covariance needs checking in the full text.

This is for readers who already work with fermionic covariance matrices in quantum chemistry or condensed-matter simulations and want a concrete number for non-Gaussian content without reconstructing the full state. The constructions are new enough and the examples concrete enough that a serious referee should see it.

Referee Report

2 major / 2 minor

Summary. The paper develops a convex resource theory for fermionic non-Gaussianity and proposes two families of computable measures for pure fermionic states derived from the Williamson normal form of the covariance matrix. The first family consists of occupation-number entropies given by the Tsallis-α entropy of the occupation numbers; the authors prove that one specific member (Tsallis-2) is monotonic under Gaussian protocols and hence a convex resource monotone that lower-bounds the number of non-Gaussian gates required for state preparation. The second family consists of natural-orbital participation entropies given by the Rényi-α entropy of the squared amplitudes of the state in the natural-orbital basis; these are shown to upper-bound classical simulation cost. The measures are evaluated on stabilizer states, translation-invariant states, random SWAP-doped matchgate circuits, and the bond-modulated XXZ model.

Significance. If the central monotonicity result holds, the work supplies the first explicitly computable convex monotones for fermionic magic that are directly obtainable from the covariance matrix, together with operational interpretations (gate lower bound and simulation-cost upper bound). The unification of resource-theoretic, condensed-matter, and quantum-chemistry perspectives, together with the concrete analyses of physically relevant states, would make the framework immediately usable for assessing simulability and complexity in fermionic many-body systems.

major comments (2)

[§4] §4 (proof of monotonicity): The claim that the Tsallis-2 occupation-number entropy is a convex resource monotone rests on the covariance matrix (via its Williamson eigenvalues) being a sufficient statistic. Because a pure fermionic state is not uniquely determined by its covariance matrix, two states can share identical occupation numbers yet differ in higher-order correlators. The proof must therefore demonstrate that any Gaussian protocol (matchgate circuit) cannot increase the underlying non-Gaussianity resource while leaving the measure unchanged; the manuscript does not explicitly address this case.
[Abstract, §3.2] Abstract and §3.2: The statement that the Tsallis-2 measure 'consequently lower bounds the number of non-Gaussian gates needed for state preparation' assumes that the resource cost is fully captured by the occupation-number entropy. If states with identical covariance but different higher-order magic require different numbers of non-Gaussian gates, the bound remains valid but its tightness and operational meaning require explicit justification or a counter-example check.

minor comments (2)

[§2] Notation for the Williamson eigenvalues and occupation numbers should be unified across §2 and §3 to avoid ambiguity between the covariance-matrix spectrum and the one-body reduced-density-matrix eigenvalues.
[§5] Figure captions for the XXZ-model and SWAP-doped-circuit plots should state the system size, boundary conditions, and exact parameter values used, as these affect the reported entropy values.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for their careful reading and constructive comments on our manuscript. We address each major comment point by point below, providing the strongest honest defense of the results while acknowledging where additional clarification is warranted.

read point-by-point responses

Referee: [§4] §4 (proof of monotonicity): The claim that the Tsallis-2 occupation-number entropy is a convex resource monotone rests on the covariance matrix (via its Williamson eigenvalues) being a sufficient statistic. Because a pure fermionic state is not uniquely determined by its covariance matrix, two states can share identical occupation numbers yet differ in higher-order correlators. The proof must therefore demonstrate that any Gaussian protocol (matchgate circuit) cannot increase the underlying non-Gaussianity resource while leaving the measure unchanged; the manuscript does not explicitly address this case.

Authors: We agree that the manuscript would benefit from an explicit remark on this point. The proof in §4 proceeds by showing that Gaussian operations (matchgate circuits) induce orthogonal transformations on the Majorana covariance matrix, which in the Williamson basis act as permutations and sign flips on the eigenvalues (occupation numbers). The Tsallis-2 entropy is Schur-concave, hence non-increasing under these transformations, establishing monotonicity under free operations. Because the free operations depend only on the covariance matrix and cannot access or create higher-order correlators, states sharing the same covariance matrix necessarily share the same value of the monotone; any apparent difference in higher-order magic is invisible to the free operations and therefore does not affect the resource-theoretic accounting. We will add a clarifying paragraph in §4 making this reasoning explicit. revision: partial
Referee: [Abstract, §3.2] Abstract and §3.2: The statement that the Tsallis-2 measure 'consequently lower bounds the number of non-Gaussian gates needed for state preparation' assumes that the resource cost is fully captured by the occupation-number entropy. If states with identical covariance but different higher-order magic require different numbers of non-Gaussian gates, the bound remains valid but its tightness and operational meaning require explicit justification or a counter-example check.

Authors: The lower-bound claim follows directly from the definition of a convex resource monotone: the minimal number of non-Gaussian gates required to prepare a state is at least the value of any monotone. This inequality holds irrespective of whether two states with identical covariance matrices possess different higher-order correlators; the monotone supplies a valid (if possibly loose) lower bound based solely on the covariance. We will revise the abstract and §3.2 to state explicitly that the bound is derived from the Tsallis-2 monotone within the resource theory and is not asserted to be tight for every state. revision: partial

Circularity Check

0 steps flagged

No circularity: definitions and monotonicity proof are independent of inputs

full rationale

The paper defines the occupation-number entropies explicitly as the Tsallis-α entropy applied to the occupation numbers extracted from the Williamson normal form of the covariance matrix, then separately states that monotonicity under Gaussian protocols is proved for one member. This is a standard resource-theoretic construction rather than a self-definitional loop, fitted parameter renamed as prediction, or load-bearing self-citation. No equations or claims in the abstract reduce the central result to its own inputs by construction; the covariance-to-occupation mapping is a standard fact for pure fermionic states and does not presuppose the monotonicity result. The derivation is therefore self-contained.

Axiom & Free-Parameter Ledger

0 free parameters · 0 axioms · 0 invented entities

Abstract-only review provides no explicit free parameters, axioms, or invented entities; all quantities are stated to derive from the standard covariance matrix of pure states.

pith-pipeline@v0.9.1-grok · 5869 in / 1111 out tokens · 39441 ms · 2026-07-03T12:17:26.103493+00:00 · methodology

0 comments

read the original abstract

Fermionic non-Gaussianity, or fermionic magic, is a key resource underlying the computational complexity of fermionic quantum systems, yet tractable and operationally meaningful ways to quantify it remain limited. We address this challenge by developing a convex resource theory of fermionic non-Gaussianity and introducing two families of computable quantities for pure fermionic states, both derived from the Williamson normal form of the covariance matrix. The first family, occupation number entropies, is defined as the Tsallis-$\alpha$ entropy of the occupation numbers. We prove that one member of this family is monotonic under Gaussian protocols, establishing it as a computable convex resource monotone. It consequently lower bounds the number of non-Gaussian gates needed for state preparation. The second family, natural-orbital participation entropies, is given by the R\'enyi-$\alpha$ entropy of the squared amplitudes of the state in the natural-orbital basis, defined by the eigenvectors of the covariance matrix. They quantify state compressibility in this basis and thus upper bound the classical simulation cost in an orthonormal Gaussian basis. We analyze both families for stabilizer and translation-invariant states, where they simplify and reveal additional structure. We further study representative examples, including random SWAP-doped matchgate circuits and the bond-modulated XXZ model, highlighting the role of non-Gaussianity in many-body phenomena. Our work establishes a resource-theoretic framework for computable fermionic non-Gaussianity that unifies notions arising across quantum information, condensed-matter physics, and quantum chemistry, opening new directions for studying the complexity of quantum many-body systems and providing practical tools to assess the classical simulability of fermionic states relevant for quantum advantage.

Figures

Figures reproduced from arXiv: 2607.02242 by Barbara Kraus, Bernhard Jobst, Frank Pollmann, Marc Langer, Poetri Sonya Tarabunga, Ra\'ul Morral-Yepes, Sheng-Hsuan Lin.

**Figure 1.** Figure 1: FIG. 1. Overview of the two families of computable measures of fermionic non-Gaussianity introduced in this work, both built [PITH_FULL_IMAGE:figures/full_fig_p004_1.png] view at source ↗

**Figure 2.** Figure 2: FIG. 2. Comparison of the occupation number entropy [PITH_FULL_IMAGE:figures/full_fig_p021_2.png] view at source ↗

**Figure 3.** Figure 3: In the first setup, shown in Fig. 3a, each time step consists of applying a random FGU to the entire system, followed by a SWAP gate. While this global setup is less physical than a local circuit, it serves as an analytically tractable model which provides a first, illustrative example of how non-Gaussianity grows in random circuits. The second setup, depicted in Fig. 3b, involves a brickwall evolution c… view at source ↗

**Figure 4.** Figure 4: FIG. 4. Time evolution of the averaged occupation num [PITH_FULL_IMAGE:figures/full_fig_p024_4.png] view at source ↗

**Figure 5.** Figure 5: FIG. 5. Occupation number entropy density [PITH_FULL_IMAGE:figures/full_fig_p025_5.png] view at source ↗

**Figure 7.** Figure 7: FIG. 7. Convergence of the occupation number entropy [PITH_FULL_IMAGE:figures/full_fig_p026_7.png] view at source ↗

**Figure 9.** Figure 9: FIG. 9. MPS algorithm to construct the matrix [PITH_FULL_IMAGE:figures/full_fig_p037_9.png] view at source ↗

**Figure 10.** Figure 10: FIG. 10. Evolution of the upper bound of [PITH_FULL_IMAGE:figures/full_fig_p046_10.png] view at source ↗

**Figure 11.** Figure 11: FIG. 11. Occupation number entropy density [PITH_FULL_IMAGE:figures/full_fig_p049_11.png] view at source ↗

**Figure 12.** Figure 12: FIG. 12. The subleading term of the occupation number en [PITH_FULL_IMAGE:figures/full_fig_p050_12.png] view at source ↗

**Figure 13.** Figure 13: FIG. 13. Convergence of the occupation number entropy den [PITH_FULL_IMAGE:figures/full_fig_p050_13.png] view at source ↗

Review history (2 revisions) →

discussion (0)

Forward citations

Cited by 2 Pith papers

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

Learning the closest Slater determinant
quant-ph 2026-07 conditional novelty 7.0

First provable algorithms with complexity bounds for learning the closest Slater determinant to an arbitrary fermionic state, plus a sharp 2/3 fidelity threshold in the optimization landscape.
Fermionic entropy: an efficiently measurable strong monotone for non-Gaussianity
quant-ph 2026-07 conditional novelty 6.0

Fermionic entropy is proven to be a strong, efficiently measurable monotone for fermionic non-Gaussianity, yielding a Theta(n) doping lower bound for matchgate-based approximate state designs.