Pith. sign in

REVIEW 4 minor 40 references

Two-copy distillability of one-copy-undistillable negative-partial-transpose states in every dimension

T0 review · 0 major / 4 minor · reviewed 2026-08-14 · deepseek-v4-flash

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

Pith's one-line read Inside a canonical family of entangled states, a state that cannot be distilled from one copy is shown to be distillable from two copies in every local dimension d≥3.

desk verdict A clean, explicit counterexample to the DiVincenzo et al. finite-copy conjecture: point C is one-copy undistillable but two-copy distillable in every dimension, with a uniform tight-frame witness. read the letter →

arxiv 2608.08836 v1 pith:CZVZESOG submitted 2026-08-09 quant-ph

Gelo Noel M. Tabia , Kai-Siang Chen , Min-Hsiu Hsieh This is my paper · ORCID
classification quant-ph
keywords entanglementdistillationnegativepartialtransposetwo-copydistillabilityone-copyundistillabilitySchmidt-rank-twowitnessequal-normtightframecanonicalNPTfamilyWernerstates
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

Entanglement distillation asks how many noisy copies of a quantum state are needed to extract pure entanglement. A long-standing open question is whether a state with a negative partial transpose (NPT) that cannot be distilled from one copy becomes distillable from finitely many copies. This paper answers the question within the canonical two-parameter family introduced as a symmetry-reduced testbed for exactly this problem: a distinguished state in its one-copy-undistillable region is already two-copy distillable in every local dimension d≥3. The authors give an explicit Schmidt-rank-two two-copy witness, built uniformly for all dimensions from an equal-norm tight frame, with a closed-form negative expectation value. Because the canonical reduction is many-to-one, the result extends to every NPT state that stochastic local operations can map to this distinguished point and disproves the conjecture that the whole one-copy-undistillable region remains undistillable for any number of copies.

What carries the argument

The load-bearing object is a uniform equal-norm tight-frame construction of the two-copy witness $|\Psi_d\rangle$. Two vectors $|x_1\rangle,|x_2\rangle$ spanning the $(d-1)$-dimensional subspace orthogonal to $|0\rangle$ are chosen so that $\langle x_r|x_s\rangle=\tfrac12\delta_{rs}$ and each coordinate has total weight $1/(d-1)$; an equally spaced real frame realises this for every $n=d-1\ge 2$. On each local two-copy space the construction takes symmetric modes $|u_r\rangle=|0\rangle|x_r\rangle+|x_r\rangle|0\rangle$ and antisymmetric modes $|v_r\rangle=|0\rangle|x_r\rangle-|x_r\rangle|0\rangle$, and pairs them as $|\Psi_d\rangle=(|u_1\rangle|v_1\rangle+|u_2\rangle|v_2\rangle)/\sqrt{2}$. The opposite copy parities make the coefficient matrix odd under simultaneous copy swap, forcing the total trace to vanish, while the equal-norm property minimizes the unavoidable positive diagonal contribution. The result is that the four expectation values of $\Delta\otimes\Delta$, $\Delta\otimes(I-\Omega)$, $(I-\Omega)\otimes\Delta$, and $(I-\Omega)\otimes(I-\Omega)$ take the fixed values $1/(d-1)$, $-1/4$, $-1/4$, and $-1/2$, giving the closed-form negative two-copy expectation at point C.

What would settle it

At $d=3$, Eq. (29) gives $\langle\Psi_3|(\rho^\Gamma_{C,3})^{\otimes 2}|\Psi_3\rangle=-1/882$ for the explicit qutrit vector in Eq. (B16); evaluating this expectation by exact arithmetic settles the theorem, and any value other than $-1/882$ would show the construction or the partial-transpose formula is wrong.

Watch

Extended reading notes

Core claim

The paper's central claim is Theorem 1: for every d≥3, the canonical state $\rho_{C,d}$ at point C is one-copy undistillable but two-copy distillable. The one-copy part follows from a decomposition in which the partial transpose of $\rho_{C,d}$ is a positive combination of the partial transpose of the one-copy-undistillable point G and positive semidefinite projectors. The two-copy part is certified directly: a single normalized vector $|\Psi_d\rangle$ of Schmidt rank exactly two across $(A_1A_2):(B_1B_2)$ satisfies $\langle\Psi_d|(\rho^\Gamma_{C,d})^{\otimes 2}|\Psi_d\rangle = -\frac{d-2}{d^2(d-1)(3d-2)^2} < 0$, which by the Schmidt-rank-two criterion proves two-copy distillability. The same witness also gives negative expectation on an open hyperbolic lobe around C, yielding explicit two-copy-distillable segments along CB and CG, and separate three-copy witnesses enlarge the certified region further. In the neighbouring triangle BGK, by contrast, recent results on Werner states combined with the propagation lemma certify two-copy undistillability, so the same one-copy-undistillable region contains states whose two-copy behaviour is provably opposite.

Load-bearing premise

The one-copy-undistillable half of Theorem 1 is not re-derived in this paper; it uses the earlier result that point G is one-copy undistillable through the decomposition in Eq. (12), so if that external result or its constants were wrong, point C would lose its advertised status.

Editorial extensions

If this is right

  • Point C becomes a certified counterexample in every dimension $d\ge 3$ to the conjecture that the whole region BCGK remains undistillable.
  • Every NPT state that stochastic local operations can map to $\rho_{C,d}$ inherits two-copy distillability, since the canonical reduction is many-to-one.
  • The fixed two-copy witness certifies an open lobe around C in the canonical parameter plane, including explicit finite segments along CB and CG with closed-form cutoff parameters.
  • Separately constructed three-copy witnesses improve the certified region for $d=3$ and $d\ge 4$, giving analytic inner bounds on the distillable set.
  • The triangle BGK is two-copy undistillable, so the canonical family contains rigorously certified states of both two-copy behaviours separated by an unresolved region.

Reading between the lines

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

  • A testable extension: the parity-and-tight-frame mechanism may transfer to higher-copy settings, so comparable closed-form witnesses could exist for three or more copies at other stages of the canonical family.
  • Since the canonical reduction collapses many states to one representative, the one-copy/two-copy gap is likely to appear in a broad stochastic-LOCC equivalence class, not only at the symmetric representative.
  • Optimising the Schmidt-rank-two vector in the unresolved part of BCG is a concrete next step: if the negative expectation reaches the boundary BG, it would show the certified two-copy lobe connects all the way to the two-copy-undistillable triangle.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

0 major / 4 minor

Summary. The paper studies the canonical two-parameter family of negative-partial-transpose (NPT) states introduced by DiVincenzo et al. and proves that, for every local dimension d ≥ 3, the distinguished state ρ_{C,d} at point C is one-copy undistillable but two-copy distillable. The two-copy certificate is an explicit Schmidt-rank-two vector |Ψ_d⟩ built from an equal-norm tight frame; Lemma 1, proved in Appendix A, gives its expectation values, and Eqs. (28)–(29) combine them into the negative value −(d−2)/(d^2(d−1)(3d−2)^2). The same witness is then used to certify an open region of two-copy-distillable states around C, and separate constructions provide three-copy witnesses that enlarge the certified region in some directions. The paper also contrasts this with the recently established two-copy undistillability of the neighboring triangle BGK, which follows from Werner-state results together with the DiVincenzo et al. propagation lemma.

Significance. If it holds, the result is significant: it disproves the conjecture that the entire one-copy-undistillable region BCGK of the canonical family remains undistillable for arbitrarily many copies, and it does so uniformly in every local dimension. The main certificate is transparent and checkable: the Schmidt-rank-two vector is explicit, the expectation values in Lemma 1 are derived from a Gram-operator calculation in Appendix A, and no parameters are fitted to force negativity. The one-copy half of Theorem 1 is imported from the standard DiVincenzo et al. result through the decomposition in Eq. (12), which the paper states clearly; this is an external but well-established input. The paper is also careful to label the two-copy lobe and the three-copy regions as witness-dependent inner bounds, not as optimized distillability boundaries. The contrast with the rigorously two-copy-undistillable triangle BGK makes the separation within one symmetry-reduced family concrete and compelling. The contribution is a strong, explicit step in a long-standing open problem.

minor comments (4)
  1. [§II.C, Eq. (12)] The decomposition in Eq. (12) is the bridge that imports one-copy undistillability from point G; a one-line verification of the coefficients, or an explicit pointer to the corresponding derivation in Ref. [8], would make the paper more self-contained.
  2. [§III.B and Abstract] The phrase "open two-copy-distillable neighborhood" around C should specify "open in the physical parameter region," because C lies on the boundary c = 0 of the physical triangle and the certified lobe is not open in the full (b,c) plane.
  3. [§III.C] In the sentence "Every two-copy certificate also yields a three-copy certificate by tensoring the corresponding Schmidt-rank-two witness with a suitable product vector on the third copy," the word "suitable" should be made explicit: the product vector must be chosen so that its one-copy expectation under ρ^Γ is positive, otherwise tensoring two negative expectations would not give a negative three-copy expectation.
  4. [Figure 2] The caption states that the panels use affine schematic coordinates and are not on a common physical scale; adding the precise defining equations for the displayed regions (e.g., Eqs. (C4), (D21), and (D26)) would help readers connect the schematic to the analytic results.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the two-copy certificate is computed from an explicit equal-norm tight frame, and the one-copy half rests on independent external results.

full rationale

The central claim, Theorem 1, has two parts. The two-copy part is fully self-contained. The witness |Psi_d> is explicitly defined in Eq. (17) from the frame vectors in Eqs. (21)-(22), and Lemma 1 derives the four expectation values (24)-(27) directly from the Gram-operator identities K_jj = 1/m, tr K = 1, and tr K^2 = 1/2, which follow from the frame construction. Substituting these into Q_d^otimes 2 gives Eq. (28) and then the negative expectation in Eq. (29); no parameter is fitted and no target negativity is imposed. The same computed witness defines the region R_d^(2) through the explicit quadratic w_{C,d}(b,c) in Eq. (C4), and the segment cutoffs in Eqs. (32)-(33) are roots of that polynomial, not fitted quantities. The three-copy witnesses in Appendix D are likewise explicit vectors with directly evaluated expectations. The one-copy part is inherited from the known one-copy undistillability of the Werner state G established by DiVincenzo et al. [8], an external result with independent content. The decomposition in Eq. (12) expresses rho^Gamma_C,d as a positive combination of rho^Gamma_G,d and a positive semidefinite operator, so point C inherits the property; this is a mathematical reduction to an external theorem, not a self-citation chain. The propagation that BGK is two-copy undistillable also uses the cited DiVincenzo et al. propagation lemma together with recent Werner-state results [16-19] published by other groups. No equation is defined in terms of the quantity it purports to derive, no fitted input is relabeled as a prediction, and no load-bearing step is justified solely by the present authors' prior work. The AI-use disclosure is a process note and does not affect the logical derivation. Accordingly, no circularity is present.

Assumptions & free parameters 1 free parameters · 3 assumptions · 0 invented entities

The only free parameter is the variational weight in the three-copy witness, which plays no role in Theorem 1. The central derivation introduces no fitted constants: the equal-norm tight frame is explicit, and the negativity in Eq. (29) is computed. The axioms are standard theorems or external results cited from the literature; none of them defines the target result.

free parameters (1)
  • lambda (three-copy frame weight) = lambda_d = 2(d-2)(5d^2-8d+4)/(13d^3-34d^2+28d-8)
    Variational parameter in the three-copy witness |Theta_{d,3}(lambda)> of Appendix D, optimized to minimize the expectation at point C. It is not fitted to data and does not affect the central two-copy theorem; it is introduced to make the three-copy certificate negative.
assumptions (3)
  • domain assumption The finite-copy distillability criterion: k-copy distillability iff there exists a Schmidt-rank-two vector |psi> with <psi|(rho^Gamma)^{otimes k}|psi> < 0.
    Invoked in Section II.A, Eq. (1), and used to certify distillability from explicit witnesses. Standard theorem from the cited literature [2,3,21], not re-derived.
  • domain assumption The Werner point G is one-copy undistillable for every d.
    Used in Eq. (12) to conclude that rho^Gamma_{C,d} is nonnegative on all Schmidt-rank-at-most-two vectors, giving the one-copy half of Theorem 1. Taken from DiVincenzo et al. [8].
  • domain assumption Recent results stating that one-copy and two-copy distillability thresholds coincide for Werner states, and the DiVincenzo propagation lemma.
    Used in Section II.B to certify that the triangle BGK is two-copy undistillable. This is contextual for the comparison claim, not required for the central counterexample.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Two-copy distillability of one-copy-undistillable negative-partial-transpose states in every dimension." pith.science (2026). https://pith.science/paper/CZVZESOG

@misc{pith2026260808836,
  author       = {Pith},
  title        = {Pith review of: Two-copy distillability of one-copy-undistillable negative-partial-transpose states in every dimension},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/CZVZESOG}},
  note         = {Machine review of arXiv:2608.08836}
}
abstract

Whether negative-partial-transpose (NPT) states that are undistillable from one copy become distillable from finitely many copies remains a basic open problem in entanglement theory. We study the canonical two-parameter family of DiVincenzo \textit{et al.}, introduced as a symmetry-reduced testbed for this question. We prove that a distinguished one-copy-undistillable state in this family is already two-copy distillable in every local dimension $d\geq 3$. A uniform equal-norm tight-frame construction gives explicit Schmidt-rank-two certificates in every dimension, thereby disproving the conjecture that the entire one-copy-undistillable region of the canonical family remains undistillable for arbitrarily many copies. The same witnesses certify an open two-copy-distillable neighborhood around the counterexample, while separately constructed three-copy witnesses enlarge the inner bounds for the distillable region in the surrounding parameter space. In contrast, recent results for Werner states, together with the propagation argument of DiVincenzo \textit{et al.}, establish a neighboring region of one-copy-undistillable states that remains two-copy undistillable. Thus a single symmetry-reduced family contains rigorously certified states with opposite two-copy behavior, separated by a substantial region whose finite-copy distillability remains unresolved.

Figures

Figures reproduced from arXiv: 2608.08836 by the authors.

Figure 1
Figure 1. FIG. 1. Parameter plane for canonical NPT states in [PITH_FULL_IMAGE:figures/full_fig_p001_1.png] view at source ↗
Figure 2
Figure 2. FIG. 2. Schematic fixed-witness regions within [PITH_FULL_IMAGE:figures/full_fig_p005_2.png] view at source ↗

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

40 extracted references · 34 canonical work pages

  1. [1]

    Peres, Separability criterion for density matrices, Phys

    A. Peres, Separability criterion for density matrices, Phys. Rev. Lett.77, 1413 (1996)

    1996

  2. [2]

    Horodecki, P

    M. Horodecki, P. Horodecki, and R. Horodecki, Mixed- state entanglement and distillation: Is there a “bound” entanglement in nature?, Phys. Rev. Lett.80, 5239 (1998)

    1998

  3. [3]

    W. Dür, J. I. Cirac, M. Lewenstein, and D. Bruß, Dis- tillability and partial transposition in bipartite systems, Phys. Rev. A61, 062313 (2000)

    work page 2000

  4. [4]

    Clarisse, The distillability problem revisited, Quantum Inf

    L. Clarisse, The distillability problem revisited, Quantum Inf. Comput.6, 539–560 (2006)

    work page 2006

  5. [5]

    Horodecki, P

    R. Horodecki, P. Horodecki, M. Horodecki, and K. Horodecki, Quantum entanglement, Rev. Mod. Phys. 81, 865 (2009)

    2009

  6. [6]

    Horodecki, L

    P. Horodecki, L. Rudnicki, and K. Życzkowski, Five open problems in quantum information theory, PRX Quantum 3, 010101 (2022)

    work page 2022

  7. [7]

    J.Watrous,Manycopiesmayberequiredforentanglement distillation, Phys. Rev. Lett.93, 010502 (2004)

    work page 2004

  8. [8]

    D. P. DiVincenzo, P. W. Shor, J. A. Smolin, B. M. Terhal, and A. V. Thapliyal, Evidence for bound entangled states with negative partial transpose, Phys. Rev. A61, 062312 (2000)

    work page 2000

Show all 40 references
  1. [9]

    R. O. Vianna and A. C. Doherty, Distillability of Werner states using entanglement witnesses and robust semidefi- nite programs, Phys. Rev. A74, 052306 (2006)

    2006

  2. [10]

    Pankowski, M

    Ł. Pankowski, M. Piani, M. Horodecki, and P. Horodecki, A few steps more towards NPT bound entanglement, IEEE Transactions on Information Theory56, 4085 (2010)

    2010

  3. [11]

    D. Ž. Ðoković, On two-distillable Werner states, Entropy 18, 216 (2016)

    2016

  4. [12]

    L. Qian, L. Chen, D. Chu, and Y. Shen, A matrix inequal- ity for entanglement distillation problem, Linear Algebra and its Applications616, 139 (2021)

    2021

  5. [13]

    S.-Y. Qi, G. Gupur, Y.-C. Wu, and G.-P. Guo, Nonpositive-transpose entanglement and bound entan- glement: From distillability sets to inequalities and multi- variable insights, Phys. Rev. A110, 012406 (2024)

    2024

  6. [14]

    Costa Rico, New partial trace inequalities and distilla- bility of werner states, Letters in Mathematical Physics 115, 47 (2025)

    P. Costa Rico, New partial trace inequalities and distilla- bility of werner states, Letters in Mathematical Physics 115, 47 (2025)

    2025

  7. [15]

    Costa Rico and M

    P. Costa Rico and M. M. Wolf, Partial trace relations beyond normal matrices (2025), arXiv:2507.18278 [quant- ph]

    2025 arXiv

  8. [16]

    J. Fu, L. Gao, and S.-J. Park, A solution to 2-copy distil- lability of Werner states (2026), arXiv:2607.21367 [quant- ph]

    2026 arXiv

  9. [17]

    Song and L

    Z. Song and L. Chen, A partial-trace matrix inequality and Werner-state distillability (2026), arXiv:2607.23416 [quant-ph]

    2026 arXiv

  10. [18]

    T. C. Fraser, F. Huber, B. Pozsgay, and I. Vona, On the two-copy distillability of Werner states and a new partial trace inequality (2026), arXiv:2607.24309 [quant-ph]

    2026 arXiv

  11. [19]

    Bharti, R

    K. Bharti, R. Gajjala, and T. Haug, Two-copy nondis- tillability of Werner states: sharp partial-trace inequali- ties and finite-copy extensions (2026), arXiv:2607.24479 [quant-ph]

    2026 arXiv

  12. [20]

    Wu and Q

    T. Wu and Q. Zou, Sharp Plücker geometry for three-copy Werner distillation (2026), arXiv:2608.02647 [quant-ph]

    2026 arXiv

  13. [21]

    Horodecki, P

    M. Horodecki, P. Horodecki, and R. Horodecki, Insepa- rable two spin-1 2 density matrices can be distilled to a singlet form, Phys. Rev. Lett.78, 574 (1997)

    1997

  14. [22]

    Horodecki and P

    M. Horodecki and P. Horodecki, Reduction criterion of separability and limits for a class of distillation protocols, Phys. Rev. A59, 4206 (1999)

    1999

  15. [23]

    R. F. Werner, Quantum states with Einstein–Podolsky– Rosen correlations admitting a hidden-variable model, Phys. Rev. A40, 4277 (1989)

    1989

  16. [24]

    Li, X.-X

    J.-Y. Li, X.-X. Fang, T. Zhang, G. N. M. Tabia, H. Lu, and Y.-C. Liang, Activating hidden teleportation power: Theory and experiment, Phys. Rev. Research3, 023045 (2021)

    2021

  17. [25]

    Horodecki, P

    M. Horodecki, P. Horodecki, and R. Horodecki, General teleportation channel, singlet fraction, and quasidistilla- tion, Phys. Rev. A60, 1888 (1999)

    1999

  18. [26]

    Verstraete and H

    F. Verstraete and H. Verschelde, Optimal teleportation with a mixed state of two qubits, Phys. Rev. Lett.90, 097901 (2003). 7 Appendix A: Coefficient-matrix motivation and equal-norm-frame expectation values

    2003

  19. [27]

    With this convention, the Schmidt rank of |Ψ⟩equals the matrix rank ofM, and normalization is equivalent to∥M∥F = 1

    Coefficient-matrix form of the point-Cterms Let |Ψ⟩= vec r(M) := d−1X a,j,b,k=0 M(a,j),(b,k) |a⟩A1 |j⟩A2 |b⟩B1 |k⟩B2 ,(A1) where M is the d2 ×d 2 coefficient matrix of|Ψ⟩ across( A1A2) : (B1B2). With this convention, the Schmidt rank of |Ψ⟩equals the matrix rank ofM, and norma...

  20. [28]

    Evaluation for the equal-norm frame Putm=d−1and define the Gram operator K= 2X r=1 |xr⟩⟨xr|(A17) onspan{|1⟩, . . . ,|m⟩}. Extend its matrix by settingK0j =K j0 = 0. The frame identities (23) imply Kjj = 1 m ,trK= 1,trK 2 = 1 2 .(A18) Write the coefficients of|Ψd⟩in the orderin...

  21. [29]

    for someD≥2, there exist local filters A:H ⊗k A − →CD, B:H ⊗k B − →CD,(B3) with nonzero success probability such that the normalized output σA,B = (A⊗B)ρ ⊗k(A† ⊗B †) tr[(A⊗B)ρ ⊗k(A† ⊗B †)] (B4) satisfiesF D(σA,B)>1/D

  22. [30]

    the output dimension can be chosen asD= 2, withF2(σA,B)>1/2. Proof. If ρ is k-copy distillable, local rank-two maps extract an entangled two-qubit state fromρ⊗k with nonzero probability. Every entangled two-qubit state can subsequently be filtered into a state with fully entan...

  23. [31]

    The all-dimensional expectation formula gives ⟨Ψ3| ρΓ C,3 ⊗2 |Ψ3⟩=− 1 882 <0.(B18) The filters below implement precisely these symmetric and antisymmetric frame modes

    Qutrit specialization of the uniform witness Ford= 3, the equal-norm frame consists of the two orthogonal coordinate directions |x1⟩= |1⟩√ 2 ,|x 2⟩= |2⟩√ 2 .(B13) Thus |u1⟩= |01⟩+|10⟩√ 2 ,|u 2⟩= |02⟩+|20⟩√ 2 ,(B14) |v1⟩= |01⟩ − |10⟩√ 2 ,|v 2⟩= |02⟩ − |20⟩√ 2 ,(B15) and the uni...

  24. [32]

    In particular, their rows are orthonormal, so they define valid stochastic local filters

    Exact qutrit filters On each local two-qutrit space, use the ordered basis {|00⟩,|01⟩,|02⟩,|10⟩,|11⟩,|12⟩,|20⟩,|21⟩,|22⟩}.(B19) DefineA 3, B3 :C 9 →C 2 by A3 = 1√ 2 0 1 0 1 0 0 0 0 0 0 0 1 0 0 0 1 0 0 ,(B20) 12 and B3 = 1√ 2 0 0−1 0 0 0 1 0 0 0 1 0−1 0 0 0 0 0 .(B21) The rows ...

  25. [33]

    Hidden teleportation power of pointC The nonzero eigenvalues ofρC,d are1 /(3d− 2)on the diagonal subspace and4/[d(3d− 2)]on the antisymmetric subspace. Therefore ∥ρC,d∥∞ = 1 3d−2 max 1, 4 d < 1 d (d≥3).(B33) It follows that Fd(ρC,d)< 1 d ,(B34) Fd2 (ρ⊗2 C,d)< 1 d2 .(B35) One-c...

  26. [34]

    (C4) gives w(d) CB (t) = d−2 8d2(d−1) 2(3d−2) 2 h (d−2)(3d−4)t 2 + 6(d−1)(d+ 2)t−8(d−1) i

    The segmentBC The relevant points are B= 1 d(d−1) ,0 , C= 4 d(3d−2) ,0 .(C8) Parameterize the segment fromCtoBby ρCB (t) = (1−t)ρ C,d +tρ B,d,0≤t≤1.(C9) Then b(t) = 4 d(3d−2) − d−2 d(d−1)(3d−2) t, c(t) = 0.(C10) Substitution into Eq. (C4) gives w(d) CB (t) = d−2 8d2(d−1) 2(3d−...

  27. [35]

    (C19) 15 The first factor is strictly positive on0≤t≤1

    The segmentCG The pointGhas coordinates G= 3 d(2d−1) , 1 d(2d−1) .(C15) Parameterize ρCG(t) = (1−t)ρ C,d +tρ G,d,0≤t≤1.(C16) Then b(t) = 4 d(3d−2) + d−2 d(2d−1)(3d−2) t,(C17) and c(t) = t d(2d−1) .(C18) The fixed-witness expectation factorizes as w(d) CG(t) = (d−1)t+ 2d−1 (5d2...

  28. [36]

    Geometry of the fixed-witness region Define R(2) d ={(b, c)physical:w C,d(b, c)<0}.(C23) The quadratic part of the numerator in Eq. (C4) has coefficient matrix Q(2) d =d 2(d−1) 3d−4d+ 4 d+ 4 4−d ,(C24) whose determinant is detQ (2) d =−4d 4(d−1) 2(d2 −2d+ 8)<0.(C25) The augmen...

  29. [37]

    Hence |Ξ3⟩= 1√ 2 (|y⟩A |y⟩B − |x⟩A |x⟩B)(D4) is normalized and has Schmidt rank exactly two across(A1A2A3) : (B1B2B3)

    A sparse qutrit witness On a local three-qutrit space, define |x⟩= |011⟩ − |102⟩ − |220⟩√ 3 ,|y⟩= |000⟩+|121⟩ − |212⟩√ 3 .(D3) The vectors are orthonormal. Hence |Ξ3⟩= 1√ 2 (|y⟩A |y⟩B − |x⟩A |x⟩B)(D4) is normalized and has Schmidt rank exactly two across(A1A2A3) : (B1B2B3). At...

  30. [38]

    Form≥3, introduce the real two-column frame f (m) ℓ1 = 1√m cos 2πℓ m , f (m) ℓ2 = 1√m sin 2πℓ m ,(D6) whereℓ= 0,

    A three-line frame witness ford≥4 Letd≥4. Form≥3, introduce the real two-column frame f (m) ℓ1 = 1√m cos 2πℓ m , f (m) ℓ2 = 1√m sin 2πℓ m ,(D6) whereℓ= 0, . . . , m−1. Form= 2, setf(2) 01 =f (2) 12 = 1/ √ 2and the other two entries to zero. In either case, m−1X ℓ=0 f (m) ℓr f ...

  31. [39]

    (D4), direct evaluation gives w(3) Ξ,3(b, c) :=⟨Ξ3| ρΓ b,c ⊗3 |Ξ3⟩ =− 1 324 27b3 + 243b2c−81b 2 + 405bc2 −270bc+ 36b−27c 3 −81c 2 + 36c−4

    The qutrit cubic For the fixed vector|Ξ3⟩in Eq. (D4), direct evaluation gives w(3) Ξ,3(b, c) :=⟨Ξ3| ρΓ b,c ⊗3 |Ξ3⟩ =− 1 324 27b3 + 243b2c−81b 2 + 405bc2 −270bc+ 36b−27c 3 −81c 2 + 36c−4 . (D21) Every physical qutrit canonical state withw(3) Ξ,3(b, c) < 0is three-copy distillab...

  32. [40]

    (D1) over the three copies and using the disjoint line supports together with Eq

    The cubic for the three-line witness Ford≥4, let w(3) d,λ(b, c) :=⟨Θd,3(λ)| ρΓ b,c ⊗3 |Θd,3(λ)⟩.(D25) Expanding Eq. (D1) over the three copies and using the disjoint line supports together with Eq. (D7) gives w(3) d,λ(b, c) =p3 +A d,λp2z−B d,λp2r+C d,λpz2 −D d,λpzr +E d,λpr2 +...

Pith tools

Reviewed August 14, 2026 · model on record in the stance chip above.