A low order Bargmann invariant hierarchy for set coherence
Yan-Ling Wang · 2026
Read A low order Bargmann invariant hierarchy for set coherence by Yan-Ling Wang (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2605-10003.txt.
Introduction
Wang, working alone, asks how little Bargmann data suffices to decide set coherence, a basis-independent form of quantum coherence. A family of states counts as set incoherent exactly when every member is diagonal in one shared basis. Bargmann invariants are cyclic trace products that expose relations between states without choosing a basis. The paper maps a low-order hierarchy for this decision.
For two states, second-order data settle qubits but not qutrits, while complete third-order data settle qutrits but fail already in dimension four. Fourth-order, ordering-sensitive invariants then give the first universal pairwise criterion. Applied to every unordered pair, that criterion tests any finite family completely. The result links cyclic trace invariants to the noncommutativity that blocks a common incoherent basis.
Catalog record
| CREATOR | Yan-Ling Wang |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 28 KB |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2605.10003
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
A low order Bargmann invariant hierarchy for set coherence Yan-Ling Wang Abstract Set coherence is a basis-independent relational form of quantum coherence: a finite family of quantum states is set incoherent exactly when all its members are diagonal in one common basis. We determine how much low-order Bargmann data are needed to decide this property. For two states, second-order data are complete for qubits but fail for qutrits, while complete third-order data are sufficient for qutrits but fail already in dimension four. We then show that fourth-order, ordering-sensitive Bargmann invariants give the first universal pairwise criterion for set coherence. Applied to all unordered pairs, this criterion yields a complete test for arbitrary finite families. The result provides a low-order hierarchy connecting cyclic trace invariants with the noncommutativity that prevents a common incoherent basis. I Introduction Quantum coherence is a basic signature of quantum behavior and a central resource in quantum information science. In the standard resource theory, coherence is defined relative to a fixed reference basis 3 ; 23 . This basis may be selected by a Hamiltonian, a measurement device, or a computational encoding. Without such a prescribed basis, however, a single density operator has no intrinsic basis-independent coherence: it can always be diagonalized in its own eigenbasis. A nontrivial basis-independent notion appears when several states are considered at the same time. Designolle et al. introduced this relational resource under the name of set coherence 4 . Given a finite family \vec{\rho}=(\rho_{1},\ldots,\rho_{n}) of states on \mathbb{C}^{d} , each individual state can be diagonalized in some basis. The question is whether all of them can be diagonalized in the same basis. If such a common basis exists, the family is set incoherent; otherwise it is set coherent. Thus set coherence does not belong to any single member of the family. It records the incompatibility among the eigenbasis structures of the states. With full knowledge of the density matrices, deciding set coherence is straightforward. A finite family of Hermitian matrices is simultaneously diagonalizable if and only if it is pairwise commuting. The question studied here is more restrictive and more invariant in spirit: can this relational nonclassicality be decided from a prescribed list of basis-independent data, without reconstructing the states and without choosing a reference basis? The basis-independent data considered in this work are cyclic traces of products of the states. We refer to them as generalized Bargmann invariants. For a word w=i_{1}i_{2}\cdots i_{m} in the state labels, the corresponding invariant is \Delta_{w}(\vec{\rho})=\operatorname{Tr}(\rho_{i_{1}}\rho_{i_{2}}\cdots\rho_{i_{m}}). These quantities are invariant under simultaneous unitary conjugation and are therefore natural for basis-independent certification. Bargmann-type invariants and related multitime trace quantities appear in geometric phases, Kirkwood–Dirac quasiprobabilities, weak values, coherence and overlap inequalities, multiphoton indistinguishability, quantum imaginarity, quantum entanglement, and relational information between states 2 ; 13 ; 5 ; 1 ; 32 ; 20 ; 6 ; 26 ; 9 ; 25 ; 10 ; 11 ; 7 ; 14 ; 30 ; 31 ; 16 ; 15 . They are also operationally meaningful, since multivariate traces can be estimated by multi-copy, interferometric, or randomized-measurement-inspired protocols 28 ; 18 ; 22 ; 21 . A recent systematic language for such data is provided by Bargmann scenarios 29 . A scenario fixes a finite set of words and studies the attainable tuple of the corresponding Bargmann invariants. In this language, one can compare quantum-realizable tuples with those generated by simultaneously diagonal families. Points that cannot arise from a common diagonal basis witness a relational resource. This viewpoint leads to the guiding question of the paper: How low in Bargmann order can one go while still deciding set coherence? At first sight, a complete answer might be expected to require many trace polynomials. Indeed, invariant-theoretic descriptions of simultaneous unitary orbits of matrices are usually generated by large families of trace polynomials 17 ; 19 ; 8 . Set coherence is a more targeted property. It does not ask for the full unitary orbit of a family, but only whether the family is simultaneously diagonalizable. This distinction leaves open the possibility that very low-order Bargmann data may already be complete in some dimensions. The purpose of this paper is to determine how far such low-order data can go and to identify the first universal pairwise order. The paper is organized as follows. Section II introduces set coherence, and completeness of a Bargmann scenario. Section III develops the two-state hierarchy from the qubit result to the four-dimensional counterexample. Section IV proves the universal fourth-order criterion and then interprets it through the Bargmann gap. Section V concludes with possible extensions. II Preliminaries Let \mathcal{D}(\mathbb{C}^{d}) denote the set of density operators on \mathbb{C}^{d} . We write \operatorname{Tr} for the trace, \mathbb{I}_{d} for the identity operator on \mathbb{C}^{d} , and [A,B]=AB-BA for the commutator. The Hilbert–Schmidt norm is \|X\|_{2}=\sqrt{\operatorname{Tr}(X^{\dagger}X)}. Definition 1 (Set coherence, 4 ) . A finite family \vec{\rho}=(\rho_{1},\ldots,\rho_{n}) , with \rho_{i}\in\mathcal{D}(\mathbb{C}^{d}) , is set incoherent if there exists a unitary U such that all U\rho_{i}U^{\dagger} are diagonal. If no such unitary exists, the family is set coherent . Since density operators are Hermitian, the standard simultaneous diagonalization criterion gives \vec{\rho}\ \text{is set incoherent}\quad\Longleftrightarrow\quad[\rho_{i},\rho_{j}]=0\quad\text{for all }i<j. For a word w=i_{1}i_{2}\cdots i_{m} in the labels 1,\ldots,n , define the generalized Bargmann invariant \Delta_{w}(\vec{\rho})=\operatorname{Tr}(\rho_{i_{1}}\rho_{i_{2}}\cdots\rho_{i_{m}}). Cyclically equivalent words give the same invariant. Definition 2 (Bargmann scenario, 29 ) . A Bargmann scenario is a finite set \mathcal{W} of words with its defining value map \Phi_{\mathcal{W}}(\vec{\rho})=\bigl(\Delta_{w}(\vec{\rho})\bigr)_{w\in\mathcal{W}}. For fixed dimension d , let \displaystyle B_{d}(\mathcal{W}) \displaystyle=\{\Phi_{\mathcal{W}}(\vec{\rho}):\vec{\rho}\in\mathcal{D}(\mathbb{C}^{d})^{n}\}, (1) \displaystyle C_{d}(\mathcal{W}) \displaystyle=\{\Phi_{\mathcal{W}}(\vec{\rho}):\vec{\rho}\in\mathcal{D}(\mathbb{C}^{d})^{n}\ \text{is set incoherent}\}, (2) \displaystyle I_{d}(\mathcal{W}) \displaystyle=\{\Phi_{\mathcal{W}}(\vec{\rho}):\vec{\rho}\in\mathcal{D}(\mathbb{C}^{d})^{n}\ \text{is set coherent}\}. (3) We say that \mathcal{W} decides set coherence in dimension d , or is complete in dimension d , if C_{d}(\mathcal{W})\cap I_{d}(\mathcal{W})=\varnothing. Equivalently, no set-incoherent and set-coherent families in the same dimension have the same \mathcal{W} -data. III The Low-Order Two-State Hierarchy For two states (\rho,\sigma) , we use the scenarios \displaystyle\mathcal{W}_{2} \displaystyle=\{11,22,12\}, \displaystyle\mathcal{W}_{3} \displaystyle=\{11,22,12,111,222,112,122\}, \displaystyle\mathcal{W}_{4} \displaystyle=\{1122,1212\}. The hierarchy proved below is summarized in Table 1 . The entry “complete” means that the corresponding data separate commuting from noncommuting pairs in the given dimension. Table 1: Completeness of the two-state Bargmann scenarios considered in this work. Complete means C_{d}(\mathcal{W})\cap I_{d}(\mathcal{W})=\varnothing . Dimension \mathcal{W}_{2} \mathcal{W}_{3} \mathcal{W}_{4} d=2 complete complete complete d=3 incomplete complete complete d\geq 4 incomplete incomplete complete We start at second order. For \mathcal{W}_{2} , write \Phi_{\mathcal{W}_{2}}(\rho,\sigma)=(x,y,z), where x=\operatorname{Tr}\rho^{2},\qquad y=\operatorname{Tr}\sigma^{2},\qquad z=\operatorname{Tr}(\rho\sigma). These three numbers are the Hilbert–Schmidt lengths of the two states and their Hilbert–Schmidt overlap. For qubits this information is already enough. Theorem 1 (Second-order completeness for qubits) . For two qubit states, the two purities and the Hilbert–Schmidt overlap decide whether the pair admits a common incoherent basis. More explicitly, \displaystyle B_{2}(\mathcal{W}_{2}) \displaystyle=\Bigl\{(x,y,z):x,y\in[1/2,1], \displaystyle\hskip 31.29802pt(2z-1)^{2}\leq(2x-1)(2y-1)\Bigr\}, (4a) \displaystyle C_{2}(\mathcal{W}_{2}) \displaystyle=\Bigl\{(x,y,z)\in B_{2}(\mathcal{W}_{2}): \displaystyle\hskip 31.29802pt(2z-1)^{2}=(2x-1)(2y-1)\Bigr\}, (4b) \displaystyle I_{2}(\mathcal{W}_{2}) \displaystyle=\Bigl\{(x,y,z)\in B_{2}(\mathcal{W}_{2}): \displaystyle\hskip 31.29802pt(2z-1)^{2}<(2x-1)(2y-1)\Bigr\}. (4c) Hence C_{2}(\mathcal{W}_{2})\cap I_{2}(\mathcal{W}_{2})=\varnothing . Proof. Write the two states in Bloch form, \rho=\frac{\mathbb{I}_{2}+\bm{r}\cdot\bm{\sigma}}{2},\qquad\sigma=\frac{\mathbb{I}_{2}+\bm{s}\cdot\bm{\sigma}}{2}, with |\bm{r}|,|\bm{s}|\leq 1 . Then 2x-1=|\bm{r}|^{2},\qquad 2y-1=|\bm{s}|^{2},\qquad 2z-1=\bm{r}\cdot\bm{s}. The inequality is Cauchy–Schwarz. Equality holds precisely when the two Bloch vectors are linearly dependent, including the case where one vector is zero. This is exactly the condition [\rho,\sigma]=0 . ∎ The qubit result has a simple geometric reason: in the Bloch ball, commutativity of two qubit states is equivalent to collinearity of