Stabilizer entropy cones as self-dual slices of linear rank cones: a five-party gap and dependence on the local dimension

arXiv:2610.18905 · cs.IT, math.CO, math.IT, quant-ph · Submitted 2026-10-11 · Read on arXiv

Listen

Transcript

Introduction to the show: ident: Quantum Radio. Generated commentary on the latest quantum physics and condensed matter papers.

Kai: Today's paper: "Stabilizer entropy cones as self-dual slices of linear rank cones".

Mira: The information is dense, highly technical, and requires careful synthesis to construct a comprehensive summary.

Kai: First, who's behind it and why it matters.

Paper summary: Kai: So, we’re looking at this paper, "Stabilizer entropy cones as self-dual slices of linear rank cones: a five-party gap and dependence on the local dimension." What the authors are doing here is taking those entanglement entropies for multipartite pure states, specifically stabilizer states—which you know are super important for quantum error correction—and showing that they generate a convex cone.

Mira: And they’re connecting that to something classical, after applying a linear lift, the closure of this stabilizer cone is exactly the self-dual slice of a classical object defined on one more element than there are parties. It says this correspondence brings two specific features of rank inequalities into play.

Lev: So it’s linking quantum stuff to matroid theory and network coding twelve twenty-one twenty-two <ref:2610.18905#pg2,matroid theory and network coding 12, 21, 22>. That’s pretty deep stuff for a paper about entanglement cones.

Kai: The first thing they highlight is that the purifier acts as an extra variable in this structure, which already suggests something different from what we thought about the quantum linear rank cone. And they show that for five parties, the stabilizer cone is strictly smaller than the quantum linear rank cone, with which earlier evidence suggested it might coincide.

Mira: They prove this gap using an explicit six-variable inequality obtained from Shannon inequalities and one common information <ref:2610.18905#pg2>. This inequality holds for stabilizer states of every local dimension, and it defines a facet of the qubit stabilizer cone, which is violated by an extreme ray of the quantum linear rank cone for five parties.

Lev: From an engineering standpoint, if we were trying to run this on real hardware, that six-variable inequality would be a specific constraint we’d have to satisfy when mapping the stabilizer states onto our physical qubit system. It tells us exactly where that gap in performance is.

Kai: Then there’s the second feature: the characteristic of the field enters, so this cone depends on the local dimension d. They point out that for up to four parties, this dependence doesn't happen, but it does for five parties.

Mira: That means if you have more than four parties, you have to worry about whether your local dimension is prime or not when comparing different qudit systems. They show that with enough parties, the cone of Z d qudits is contained in the cone of Z d' qudits if and only if d divides d', provided there are at least two (d, d') + three parties <ref:2610.18905#pg2>.

Lev: That’s a lot of constraints on how you can embed one type of quantum system into another one when the number of parties gets large. It really limits the flexibility we have in designing error correction codes across different qudit sizes.

Kai: And they give us a concrete example: for six parties, the logical zero state of Steane's code lies outside the stabilizer cone for every odd local dimension <ref:2610.18905#pg3>. That’s a specific, known error-correcting code state that doesn't fit into this structure.

Mira: So what this all boils down to is a powerful dictionary: we translate problems about the stabilizer entanglement of n parties into questions about the representability of self-dual rank functions on n+one elements <ref:2610.18905#pg1>.

Lev: It shifts the focus away from just measuring entropy and towards whether these classical combinatorial objects can actually be constructed for a given dimension.

Kai: The overall message from this paper, "Stabilizer entropy cones as self-dual slices of linear rank cones: a five-party gap and dependence on the local dimension," is that we need to account for the purifier acting as an extra variable, and these inequalities can explicitly depend on the characteristic of the field.

Mira: This refines our understanding of stabilizer entanglement by providing this combinatorial framework with both those extra degrees of freedom and those field-dependent constraints, especially when you have five or more parties.

Lev: It's a lot of technical detail to manage, but it gives us a much clearer boundary for what’s possible in terms of how these quantum states relate to classical structures.

Kai: So, if you’re driving and thinking about your commute, this paper means that the structure of entanglement isn't as simple as it looked before; there are hidden dependencies on the number of parties and the math underneath.

Conclusion: Kai: So we’re wrapping up this look at the paper, "Stabilizer entropy cones as self-dual slices of linear rank cones: a five-party gap and dependence on the local dimension." It’s about how those entanglement entropies for stabilizer states map onto these classical combinatorial objects.

Mira: It boils down to showing that after applying a linear lift, the closure of that stabilizer cone is exactly the self-dual slice of some classical thing defined over n elements, where n is the number of parties.

Lev: And they connect this to rank functions of arrangements over GF(d), which basically means they’re mapping quantum states onto geometry and matroid theory.

Kai: I think the real meat here is that they’re highlighting two things we missed before. First, the purifier acts as an extra variable in the system, which changes how we look at these limits.

Mira: Exactly. And second, this entire structure depends on the characteristic of the field and whether you have enough parties to see it matter.

Lev: For five parties, they show a gap between the stabilizer cone and that quantum linear rank cone because of that purifier variable, even though earlier people thought they might be the same thing.

Kai: That six-variable inequality is a concrete witness to that gap; it shows how much smaller the stabilizer structure really is compared to what we expected.

Mira: And then you’ve got the dependence on the local dimension d coming into play, which isn't always present for fewer parties.

Lev: They show that if you have enough parties, say 2d plus three or more, then comparing different qudit systems just means checking if one dimension divides another <ref:2610.18905#pg1>.

Kai: It’s a big statement about how restrictive those constraints are on which quantum systems can fit together in these error correction codes.

Mira: And they give us an example with Steane's code, showing that the logical zero state doesn't even fit into the stabilizer cone for every odd local dimension.

Lev: So essentially, this paper gives us a new dictionary to translate entanglement questions into representability problems in matroid theory.

Kai: It’s a shift from just measuring entropy to checking if those classical structures can actually be built for a given setup.

Mira: The big picture is that we need to include the purifier and the field dependence when we analyze stabilizer entanglement for systems with five or more parties.

Lev: It’s important because it tells us exactly where those performance gaps are and how much flexibility we lose when moving to larger qudit systems.

Kai: Next up, we’re going to look at some of the specific bounds they set on this problem.

Xidan Song, Ruifeng Cao

School of Computer Science, Wuhan University · School of Engineering, The University of Manchester

cs.IT, math.CO, math.IT, quant-ph

Submitted: 2026-10-11

Updated: 2026-10-11

Comments: 29 pages, 4 figures, 6 tables. Data, certificates and code: doi:10.5281/zenodo.23303466. Submitted to PRX Quantum

Code: https://github.com/RuifengCao/quantum-linear-rank-cone

License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/

The gist: The information is dense, highly technical, and requires careful synthesis to construct a comprehensive summary.

Key concepts

Stabilizer States: These are specific types of quantum states crucial for quantum error correction. The paper analyzes their entanglement entropies to build a geometric structure (a cone) that describes how entangled these states can be.
Convex Cone: A convex cone is a geometric shape where if you take any two points within the shape, the line connecting them stays entirely within the shape. In this context, it represents all possible achievable entanglement vectors for stabilizer states.
Linear Rank Cone: This is a mathematical structure related to how many independent pieces of information can be extracted from a set of quantum measurements. The paper compares the stabilizer cone to this broader cone to find where they differ, particularly for five parties.

Summary

The information is dense, highly technical, and requires careful synthesis to construct a comprehensive summary.

Here is the detailed and long summary combining the provided material:

Comprehensive Summary: Stabilizer Entanglement Cones as Self-Dual Slices of Linear Rank Cones

This research investigates the structure of entanglement entropies for multipartite pure states, specifically focusing on stabilizer states, which form the backbone of quantum error correction. The core finding establishes a deep connection between these entanglement properties and classical combinatorial objects: the achievable entropy vectors generate a convex cone. Crucially, after applying a linear lift, the closure of this stabilizer cone is shown to be exactly the self-dual slice of a classical object defined over n+1 elements, where n is the number of parties.

Key Findings on Structure and Inequalities:

The Classical Correspondence: The closed cone generated by these rank functions corresponds precisely to the cone generated by the rank functions of arrangements of subspaces over GF(d) for a prime local dimension d, and subgroups of abelian groups whose exponent divides d in the general case. This correspondence links quantum information theory directly to representation theory, specifically matroids and polymatroids.
The Role of the Purifier: A significant feature highlighted is that the purifier acts as an extra variable in this structure. This leads to two primary insights:

Gap for Five Parties: For five parties, the stabilizer cone is strictly smaller than the quantum linear rank cone (QLR5). An explicit six-variable linear rank inequality, derived from Shannon inequalities and common information measures, is presented as a witness to this gap. This result directly addresses Question 1 (Q1) in the context of five parties in a negative sense.
Dependence on Characteristic: Unlike earlier evidence suggesting coincidence between the stabilizer cone and the QLR cone, the explicit six-variable inequality does depend on the characteristic of the underlying field GF(d).

Dependence on Local Dimension (d): The structure of this cone is highly sensitive to d, which represents a local dimension related to qudit systems.

Four Parties: For up to four parties, the cone structure remains independent of d (Theorem 2).
Five Parties: For five parties, the dependence on d becomes apparent. The characteristic of the field enters the inequalities, meaning the cone does not depend on d for up to four parties but does for five.
General Dependence: With enough parties, this dependency becomes universal. Specifically, for primes p < q, the cones of Z p and Z q qudits are shown to be incomparable when there are at least 2p + 3 parties (Theorem 4). Furthermore, with sufficient parties, the cone of Z d qudits is contained within that of Z d' qudits if and only if d divides d', provided there are at least 2 (d, d') + 3 parties (Theorem 5).

Specific Code Examples: The analysis extends to known quantum error-correcting codes. For instance, the logical zero state of Steane’s code is shown to lie outside the stabilizer cone for every odd local dimension (Theorem 6), highlighting limitations of the cone structure in certain code contexts.

Theoretical Implications and Dictionary:

The research establishes a powerful dictionary that translates problems in stabilizer entanglement (Stab(d) n) into questions concerning the representability of self-dual rank functions on n+1 elements. This shift allows researchers to trade questions about the structure of Stab(d) n for more tractable representability problems within matroid and polymatroid theory.

The key takeaway is that the quantum linear rank cone, as previously studied, does not fully capture these features. The current work reveals two critical aspects that are not apparent in earlier investigations:

The purifier introduces an extra variable into the system.
The resulting inequalities may explicitly depend on the characteristic of the field.

In essence, this paper refines our understanding of stabilizer entanglement by providing a precise combinatorial and algebraic framework (matroids/polymatroids) that incorporates both the extra degrees of freedom (purifier) and the field-dependent constraints, particularly for systems involving five or more parties. The work is supported by extensive computational verification using advanced AI tools, ensuring that every computational claim is backed by exact certificates or recomputations.

Improvements for AI systems

Improve entanglement analysis for stabilizer states by leveraging Theorem 1 to establish a dictionary between stabilizer entropy vectors and self-dual slices of linear rank cones. This allows systems to determine if an observed entanglement pattern is realizable by any local dimension, as the paper states: The dictionary connects stabilizer entanglement with the representation theory of matroids and polymatroids in both directions.
Develop a mechanism to distinguish between QLRn and the stabilizer cone for five parties by utilizing Theorem 3, which provides an explicit separating inequality: Stab5 ⊊ QLR5, thereby creating a rigorous test to determine if a set of entanglement measurements is realizable by stabilizer states of all local dimensions.
Implement characteristic-dependent analysis for quantum error correction by using the results from Section V, which show that the cone depends on the local dimension, allowing AI systems to predict whether a specific code state (e.g., Steane's code) will be realizable in qudits of a given dimension based on divisibility rules like Stab(p)n ⊊ Stab(d)n for every prime p dividing d.
Enhance the detection of subtle entanglement patterns by incorporating the purifier as an extra variable into inequality checks, as noted in Section IV: The purifier is an additional variable, which allows the system to distinguish between QLRn and stabilizer cones by checking inequalities that are valid for arrangements of n + 1 subspaces over GF(d), one of which belongs to the purifier.
Create a comprehensive catalog search tool based on Theorem 6, enabling the identification of known realizable entanglement patterns by searching through the catalog of extreme rays: The extreme rays of QLR5 were enumerated by adjacency decomposition [50], starting from the 59 known orbits [15] and the vector q2.