Pith. sign in

REVIEW 5 minor 6 references

No-Go Theorems for Quantum Transport Metrics from Fixed Cost Operators

T0 review · 0 major / 5 minor · reviewed 2026-08-15 · deepseek-v4-flash

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

Pith's one-line read This paper proves that no fixed state-independent Hermitian cost can make coupling-based quantum transport a metric for d≥3.

desk verdict A genuine uniform no-go theorem for fixed-cost quantum transport metrics, built on an external admissibility characterization; one repairable conjugation slip in the channel-induced section, but the main result survives. read the letter →

arxiv 2607.28358 v1 pith:UTECZ4BE submitted 2026-07-30 quant-ph

Minbo Gao , Zhengfeng Ji , Tianshi Yu This is my paper · ORCID
classification quant-ph
keywords quantumoptimaltransportmetricno-gotheoremfixedcostoperatortriangleinequalitySWAPcommutingstatesstabilized
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

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

The reading

The paper proves a uniform no-go theorem for coupling-based quantum optimal transport: in every Hilbert-space dimension d≥3, no fixed, state-independent Hermitian cost operator can make either the optimal coupling cost or its square root a metric on the full set of mixed states. The triangle inequality fails already for three mutually commuting qutrit states, and one fixed triple works, after a basis change, for every admissible cost. A separate result shows the channel-induced formulation fails even earlier: global nonnegativity together with zero self-cost forces the cost operator to be zero, so no point separation is possible for d≥2. The paper also shows that stabilizing the SWAP cost by tensoring with arbitrary ancillas does not restore the triangle inequality. Taken together, these statements close the question in the negative: fixed-cost coupling formulations do not produce quantum Wasserstein metrics on all mixed states.

What carries the argument

The load-bearing mechanism is the admissible-cost characterization, which reduces every candidate metric to a positive operator with kernel equal to the symmetric subspace. From there, the qutrit antisymmetric normal form is the key structural reduction: because every orthonormal basis of $\wedge^2\mathbb{C}^3$ is, up to phases, a wedge basis induced by a one-particle basis, any admissible qutrit cost becomes a sum of three positive weights on the antisymmetric directions $|1\wedge2\rangle,|1\wedge3\rangle,|2\wedge3\rangle$. The diagonal-state reduction then turns the quantum optimization into a classical coupling minimization with square-root terms, $T=\frac12\min_\gamma\sum_{i<j}c_{ij}(\sqrt{\gamma_{ij}}-\sqrt{\gamma_{ji}})^2$; this is the identity that carries the argument, since the square roots are generated by quantum coherence inside each $\{|ij\rangle,|ji\rangle\}$ block. For stabilization, the function $\Phi(\pi)=\sum_{i,j}\sqrt{\pi_{ij}\pi_{ji}}$ is shown to be unaffected by coarse-graining an ancilla, so tensoring with a common state does not change the SWAP cost on commuting pairs. For channel-induced couplings, the mechanism is a polarization identity: every basis tensor $|e_i\rangle\otimes|e_j\rangle$ is a linear combination of vectors $|\psi\rangle\otimes|\overline{\psi}\rangle$, so zero self-cost on pure states spans the kernel of $C$.

What would settle it

Compute, for a fixed admissible qutrit cost with weights $a,b,c>0$, the three pairwise costs on the commuting states with marginals $p=(3/4,1/4,0)$, $q=(1/4,5/8,1/8)$, and $r=(1/8,5/8,1/4)$, using $\frac12\min_\gamma\sum_{i<j}c_{ij}(\sqrt{\gamma_{ij}}-\sqrt{\gamma_{ji}})^2$. If the strict inequalities $T_C(\rho,\tau)>T_C(\rho,\sigma)+T_C(\sigma,\tau)$ and $W_C(\rho,\tau)>W_C(\rho,\sigma)+W_C(\sigma,\tau)$ fail for any positive weights, the uniform theorem is false; equivalently, finding any $d\ge3$ admissible cost whose square-root cost satisfies the triangle inequality on all mixed states would refute it.

Watch

Extended reading notes

Core claim

Under the standard coupling set—bipartite states with prescribed partial traces—a Hermitian cost $C$ is admissible precisely when $C \succeq 0$ and its kernel is the symmetric subspace $\mathrm{Sym}^2(H)$. The paper uses this characterization to compress any admissible cost to a qutrit, where a basis change brings it to the normal form $C_{a,b,c}=a|1\wedge2\rangle\langle1\wedge2|+b|1\wedge3\rangle\langle1\wedge3|+c|2\wedge3\rangle\langle2\wedge3|$ with $a,b,c>0$. For commuting diagonal states the cost reduces to $T_{C_{a,b,c}}(\mathrm{diag}\,p,\mathrm{diag}\,q)=\frac12\min_{\gamma\in\Gamma_{\mathrm{cl}}(p,q)}\sum_{i<j}c_{ij}(\sqrt{\gamma_{ij}}-\sqrt{\gamma_{ji}})^2$, where the square roots come from coherent off-diagonal blocks of the quantum coupling. The paper then exhibits one fixed commuting triple of diagonal qutrit states for which the triangle inequality fails for every positive $a,b,c$, both for $T_C$ and for $\sqrt{T_C}$. For channel-induced couplings, nonnegativity of all optimized values gives $C\succeq0$, and zero self-cost on pure states gives $C|\psi\otimes\overline{\psi}\rangle=0$ for every $\psi$; since these vectors span the whole bipartite space, $C=0$. The conclusion is that no fixed linear cost operator yields a metric on the full mixed-state space in dimension $d\ge3$, and no channel-induced cost even separates points when $d\ge2$.

Load-bearing premise

The load-bearing premise is the external characterization theorem that a Hermitian cost gives a semidistance only if it is positive semidefinite and vanishes exactly on the symmetric subspace—the two-particle states unchanged by swapping the two particles; if that characterization omits admissible costs or is wrong, the reduction to three qutrit weights would not cover all fixed-cost semidistances.

Editorial extensions

If this is right

  • For every $d\ge3$, no state-independent linear cost can make the standard coupling cost $T_C$ or its square root $W_C$ a metric on all mixed states; the threshold is sharp because in $d=2$ the cost is a multiple of the SWAP projector and $W_C$ is the usual metric on the Bloch ball.
  • The stabilized SWAP cost, despite gaining channel monotonicity and invariance under tensoring with a common ancilla, still violates the triangle inequality, so finite-dimensional stabilization cannot repair the metric property.
  • For channel-induced couplings in dimension $d\ge2$, the only operator satisfying nonnegativity and zero self-cost is $C=0$, so point separation—and hence any semidistance—is impossible.
  • The no-go is uniform rather than perturbative: one fixed commuting qutrit triple violates the triangle inequality for every admissible cost after a basis change, so the failure is not special to exact SWAP or nearby costs.
  • Metricity can survive on restricted families—on pure states the SWAP cost is a constant multiple of root infidelity and hence a metric—but not on the full mixed-state space.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • A natural next step, which the paper leaves open, is to allow state-dependent costs or multi-round adaptive plans; the uniform no-go suggests no single fixed operator can reproduce classical Wasserstein geometry on all states, so any metric formulation must change the cost rule with the states involved.
  • The deeper obstruction may be the failure of quantum gluing: the paper shows the triangle inequality itself fails, not merely the classical proof route, which points toward studying non-coupling-based constructions or higher-order coupling objects.
  • A testable extension: with a two-qutrit device implementing an estimated cost operator, one could measure the three pairwise costs on the commuting triple $p=(3/4,1/4,0)$, $q=(1/4,5/8,1/8)$, $r=(1/8,5/8,1/4)$; the predicted strict triangle violation is quantitative and should survive noise unless the cost operator becomes inadmissible.
  • It remains unclear whether the dimension threshold changes if one restricts to pure states or to other subfamilies of commuting states; the paper's proof is specific to full mixed-state space, so weaker no-go statements for such families are not implied.
Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

0 major / 5 minor

Summary. The paper proves two no-go theorems for fixed-cost couplings in quantum optimal transport. For the standard coupling set (3)-(4), Theorem 1.1 shows that for any Hermitian cost C on C^d⊗C^d, d≥3, if T_C is a semidistance, then T_C and W_C=√T_C violate the triangle inequality on a fixed triple of commuting qutrit states. The proof reduces arbitrary admissible costs to three positive weights via the external characterization [CEF˙Z23, Thm 6.3], uses the exact diagonal-state formula (24), and verifies a uniform weighted counterexample in Proposition 3.6. Corollary 1.2 transfers the violation to the stabilized SWAP cost, proving it is not a metric. For channel-induced couplings (5), Theorem 1.3 shows that global nonnegativity together with zero self-cost forces C=0, so no semidistance or metric exists for d≥2. The dimension threshold d≥3 is sharp because in dimension two all admissible costs are multiples of SWAP and the square-root cost is a metric.

Significance. If correct, the main theorem settles the open question posed by Friedland et al. by ruling out, for d≥3, all fixed linear cost operators as a source of quantum Wasserstein metrics on the full mixed-state space, thereby sharpening Miller's qutrit counterexample to a uniform statement. The paper contains several independently useful technical tools: the qutrit antisymmetric normal form (Lemma 3.3), the twirl-based proof of the diagonal-state formula (Lemma 3.4), the explicit three-state counterexample with strict inequalities for all positive weights, and the stabilization argument (Proposition 4.1). The proofs are detailed, and I verified the algebra in Proposition 3.6 and the coarse-graining argument in Proposition 4.1; the channel-induced argument in Section 5 is short and correct. The main theorem is conditional on the admissibility characterization quoted as Theorem 3.1, but this is a citation to a published result rather than an internal gap.

minor comments (5)
  1. [Section 3.1, Theorem 3.1] The proof of Theorem 1.1 uses the full "if and only if" characterization of admissible costs from [CEF˙Z23, Theorem 6.3], but the paper proves only the easy inclusion. Please add a remark or an appendix proof making explicit that the uniform reduction to (11) is conditional on the necessity direction of this external theorem; this will help readers who want to verify the coverage of all admissible costs.
  2. [Section 5, Proposition 5.3] The unique element of the channel-induced self-coupling set for a pure state |ψ⟩⟨ψ| is |bar-ψ⊗ψ⟩⟨bar-ψ⊗ψ|, with the first factor carrying the transpose, not |ψ⊗ψ⟩⟨ψ⊗ψ|. The conclusion is correct because every vector bar-ψ occurs as ψ ranges over H, but the notation should be corrected for consistency with (5).
  3. [Lemma 5.2 proof] The displayed cancellation sums are mis-stated: vanishing of the |y⟩⊗|y⟩ terms uses ∑_{k=0}^3 (-i)^k(-1)^k=0, not ∑_{k=0}^3 (-1)^k=0. The polarization identity (44) itself is correct.
  4. [Around Eq. (5) and Section 5] Please clarify the convention relating the transpose ρ^T to the entrywise conjugate |bar-ψ⟩ used in Section 5; this would remove the ambiguity in the pure-state argument and in the spanning lemma.
  5. [Theorem 1.1, after Eq. (27)] The phrase "one fixed triple" should read "one fixed triple in the basis supplied by Lemma 3.3," since the triple in (27) is specified after a C-dependent basis change; the current wording could be misread as claiming a single numerical triple works for every C.

Circularity Check

0 steps flagged · score 0.0 of 10

No circularity: the no-go theorem is derived from external characterization theorems and self-contained reductions.

full rationale

The paper's central claim (Theorem 1.1) depends on two external results: the admissible-cost characterization of [CEF˙Z23, Theorem 6.3] (quoted as Theorem 3.1) and Miller's diagonal-state formula [Mil26, Proposition 1] (reproved as Lemma 3.4). Neither is equivalent to the target no-go statement, and neither is a self-citation: the authors of the present paper are Minbo Gao, Zhengfeng Ji, and Tianshi Yu, with no overlap on [CEF˙Z23] or [Mil26]. The characterization is an independent structural theorem about which Hermitian costs make T_C a semidistance; the paper uses its necessity direction to reduce an arbitrary admissible cost to the normal form C_{a,b,c} via Lemma 3.2 and Lemma 3.3, both proved in full. The diagonal-state formula is also proved in the text via twirling, and Proposition 3.6 is a new weighted-qutrit calculation that yields the strict triangle-inequality violations for all positive a,b,c. The only self-citations ([BGWZ25], [BGK+26]) appear in the introduction as application references and are not load-bearing. No fitted parameter is relabeled as a prediction, no ansatz is smuggled in via citation, and no known result is merely renamed. Any concern about the correctness or completeness of the external characterization is a robustness or correctness risk, not circularity, because the derivation does not assume its conclusion. The proof chain is therefore self-contained modulo clearly identified external theorems.

Assumptions & free parameters 0 free parameters · 1 assumptions · 0 invented entities

The central theorem rests on the CEF˙Z23 characterization of admissible costs; without it, the reduction to positive antisymmetric operators is not justified. The rest of the proof is self-contained, with Miller's diagonal-state formula reproved in Lemma 3.4 and the qutrit normal form proved in Lemma 3.3. No free parameters are fitted and no new entities are introduced.

assumptions (1)
  • standard math Admissible cost characterization: C is Hermitian and T_C is a semidistance iff C ⪰ 0 and ker C = Sym²(H) (CEF˙Z23, Theorem 6.3).
    External theorem imported without proof. It is used in Section 3.1 to reduce an arbitrary admissible cost to a positive operator on the antisymmetric subspace. The main no-go theorem depends on this equivalence.

how reviews work

0 comments
Cite this review

Pith. "Pith review of No-Go Theorems for Quantum Transport Metrics from Fixed Cost Operators." pith.science (2026). https://pith.science/paper/UTECZ4BE

@misc{pith2026260728358,
  author       = {Pith},
  title        = {Pith review of: No-Go Theorems for Quantum Transport Metrics from Fixed Cost Operators},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/UTECZ4BE}},
  note         = {Machine review of arXiv:2607.28358}
}
abstract

Coupling-based quantum optimal transport generalizes classical optimal transport by representing transport plans as bipartite states with prescribed marginals and evaluating their cost as the expectation of a fixed Hermitian operator. Friedland et al. [Phys. Rev. Lett. 129, 110402 (2022)] conjectured that the square root of the optimal cost associated with the SWAP projector is a metric in every dimension and that this property persists for nearby quantum cost matrices. Miller [arXiv:2607.07764] disproved both conjectures by constructing explicit diagonal qutrit counterexamples. Building on his analysis, we prove a uniform no-go theorem for standard couplings. In every dimension $d\geq3$, no fixed cost operator makes either the optimal cost or its square root a metric, with violations occurring already among commuting states. The obstruction persists under stabilization of the SWAP cost. For channel-induced couplings, global nonnegativity and vanishing self-cost force the cost operator to be zero, precluding point separation when $d\geq2$. Taken together, these no-go results show that fixed-cost coupling formulations do not lead to metrics on the full quantum state space.

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

6 extracted references · 1 canonical work pages

  1. [1]

    Birkh¨ auser, Basel, 2 edition, 2008.doi:10.1007/ 978-3-7643-8722-8

    [AGS08] Luigi Ambrosio, Nicola Gigli, and Giuseppe Savar´ e.Gradient Flows: In Metric Spaces and in the Space of Probability Measures. Birkh¨ auser, Basel, 2 edition, 2008.doi:10.1007/ 978-3-7643-8722-8. 2 [BGK+26] Gilles Barthe, Minbo Gao, Jam Kabeer Ali Khan, Matthijs Muis, Ivan Renison, Keiya Sak- abe, Michael Walter, Yingte Xu, Tianshi Yu, and Li Zhou...

    work page 2008

  2. [1958]

    1 [MH22] Alexander M¨ uller-Hermes

    doi:10.1287/mnsc.5.1.1. 1 [MH22] Alexander M¨ uller-Hermes. On the Monotonicity of a Quantum Optimal Transport Cost,

    doi:10.1287/mnsc.5.1.1

  3. [2016]

    2 [Kan58] L

    doi:10.1007/s00220-015-2485-7. 2 [Kan58] L. Kantorovitch. On the translocation of masses.Management Science, 5(1):1–4, October

    doi:10.1007/s00220-015-2485-7

  4. [2019]

    2 [DPT21] Giacomo De Palma and Dario Trevisan

    URL:https://proceedings.neurips.cc/ paper_files/paper/2019/file/f35fd567065af297ae65b621e0a21ae9-Paper.pdf. 2 [DPT21] Giacomo De Palma and Dario Trevisan. Quantum Optimal Transport with Quantum Channels. Annales Henri Poincar´ e, 22(10):3199–3234, October 2021.doi:10.1007/s00023-021-01042-3. 5, 6 [FEC ˙Z22] Shmuel Friedland, Micha l Eckstein, Sam Cole, an...

    doi:10.1007/s00023-021-01042-3 2019

  5. [2022]

    3, 13 16 [Mil26] Tomasz Miller

    arXiv:2211.11713. 3, 13 16 [Mil26] Tomasz Miller. Comment on ’Quantum Monge-Kantorovich Problem and Transport Distance between Density Matrices’, July 2026.arXiv:2607.07764. 1, 2, 4, 9, 14 [PC19] Gabriel Peyr´ e and Marco Cuturi. Computational optimal transport: With applications to data science.Foundations and Trends in Machine Learning, 11(5–6):355–607,...

    arXiv 2026

  6. [2026]

    doi:10.4230/LIPIcs.LICS.2026.15

    Schloss Dagstuhl – Leibniz-Zentrum f¨ ur Informatik. doi:10.4230/LIPIcs.LICS.2026.15. 2 [BGWZ25] Gilles Barthe, Minbo Gao, Theo Wang, and Li Zhou. Complete Quantum Relational Hoare Logics from Optimal Transport Duality. In2025 40th Annual ACM/IEEE Symposium on Logic in Computer Science (LICS), pages 884–925, 2025.doi:10.1109/LICS65433.2025.00072. 2 [CEF ˙...

    arXiv 2026

Pith tools

Reviewed August 15, 2026 · model on record in the stance chip above.