Pith. sign in

REVIEW 2 major objections 5 minor 44 references

Extending quantum correlations to arbitrary distances via parallel repetition of routed Bell tests

T0 review · 2 major / 5 minor · reviewed 2026-08-16 · deepseek-v4-flash

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

Pith's one-line read Parallel repetition of routed Bell tests makes the critical detection efficiency drop exponentially to 1/2^N.

desk verdict The rBB84_N threshold is a clean, citable result, but the rCHSH_N proof has a false bound in Eq. (B17) that invalidates Proposition 2 as written. read the letter →

arxiv 2504.17621 v1 pith:VQRATIJ3 submitted 2025-04-24 quant-ph

Anubhav Chaturvedi , Marcin Paw{l}owski , Máté Farkas This is my paper · ORCID
classification quant-ph
keywords routedBellexperimentsparallelrepetitiondetectionefficiencynon-jointlymeasurablemeasurementsdevice-independentquantumkeydistributionself-testingCHSHinequalityBB84
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

This paper tries to establish that the detector-efficiency barrier to long-distance quantum correlations can be broken exponentially by repeating routed Bell experiments in parallel. Concretely, for any number $N$ of parallel copies, the two strategies $rBB84_N$ and $rCHSH_N$ certify that the distant device's measurements are non-jointly measurable—the property that blocks an eavesdropper from faking the statistics—whenever the distant detection efficiency exceeds $\eta^* = 1/2^N$. Because the distant device has $2^N$ measurement settings, this threshold cannot be lowered for these strategies, so the exponential decrease is optimal. If true, the result is a concrete route toward loophole-free certification and device-independent quantum key distribution over much longer distances, at the price of preparing $2^N$-dimensional entangled states and keeping the near-source devices near-perfect.

What carries the argument

The load-bearing object is the $q$-penalized $N$-product Bell inequality, $B_N(q) = B_N - (q/2^N)\sum_{y,b} p_B(b|y,1)$, a generalized long-path inequality that discounts every click at the distant device. Self-testing is the second ingredient: maximal violation of the $N$-product CHSH inequality between $A$ and $B_0$ implies, up to an isometry and a junk state, the reference strategy, yielding two consequences used throughout: Alice's marginals are uniform and Bob's remotely prepared states are Alice's reference projectors. Under a jointly measurable model at $B_1$, the $q$-penalized value becomes a sum of terms involving operators $C_{\vec b}$ built from tensor products of projections; bounding the largest eigenvalue of these operators through a norm inequality for sums of positive semidefinite operators gives the jointly measurable ceiling $(1-q)/2^N$. Comparing the ceiling with the ideal value $(1-q)\eta$ yields $\eta^* = 1/2^N$; the $rCHSH_N$ case follows the same template with two-qubit density operators replacing projections.

What would settle it

For $N=4$, evaluate the largest eigenvalue of every operator $\sum_{y: b_y \neq \varnothing} ( \bigotimes_j A^{y_j}_{b_{y,j}} - q I )$ at $q = 1/\sqrt{2}$; if any eigenvalue exceeds $1 - 1/\sqrt{2}$, the jointly-measurable bound in Proposition 1 is false.

Watch

Extended reading notes

Core claim

The central claim is that the $N$-fold parallel repeated routed strategies $rBB84_N$ and $rCHSH_N$ achieve the optimal detection-efficiency threshold $\eta^* = 1/2^N$ for certifying that the far device $B_1$ performs non-jointly measurable (NJM) measurements, assuming the source and the near devices $A$ and $B_0$ are perfect. The proof introduces $q$-penalized versions of the $N$-product BB84 and CHSH inequalities, in which every click of $B_1$ is penalized. Self-testing from the maximal violation of the $N$-product CHSH inequality between $A$ and $B_0$ forces Alice's outcomes to be uniform and fixes Bob's remotely prepared states; inserting this into a jointly measurable model of $B_1$ bounds any such model by $(1-q)/2^N$, while the ideal strategies achieve $(1-q)\eta$. Hence NJM is certified whenever $\eta > 1/2^N$, and since $B_1$ has $2^N$ settings the threshold is tight.

Load-bearing premise

The proofs require the near-source devices $A$ and $B_0$ to achieve the exact maximum quantum violation of the $N$-product CHSH inequality, because the self-testing relations used to bound jointly measurable models hold only under that perfection.

Editorial extensions

If this is right

  • Each additional parallel copy halves the detection efficiency needed to certify non-jointly measurable measurements at the distant device, from $1/2$ at $N=1$ to $1/2^N$ in general.
  • The threshold is tight for these strategies: because $B_1$ has $2^N$ settings, no certification of NJM measurements at that device can succeed at efficiency below $1/2^N$.
  • The certification step on which secure device-independent quantum key distribution relies can in principle run at this exponentially lower efficiency, although a full key-rate analysis is not part of the paper.
  • The cost of the exponential gain is exponential resources: the shared state has local dimension $2^N$, and the distant device needs $2^N$ measurement settings.
  • For $N=2$, numerical relaxations show that both parallel strategies tolerate source noise better than their single-copy counterparts, and better than a qubit-based strategy with four distant settings.

Reading between the lines

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

  • Beyond the paper: because NJM certification is the cryptographic precondition, the same $1/2^N$ law should set the threshold for positive key rates in DIQKD, but proving that requires a full finite-key analysis the paper does not perform.
  • Beyond the paper: the tightness argument suggests a general trade-off: any routed strategy giving $B_1$ $m$ settings cannot certify NJM below $1/m$, so the exponential gain is bought by exponential dimension and settings, meaning a practical sweet spot likely lies at moderate $N$.
  • Beyond the paper: the norm-bound technique, applied here to product BB84 and CHSH inequalities, may extend to other self-testable Bell inequalities; a testable prediction is that the same $1/2^N$ scaling appears for any inequality whose maximal violation self-tests product measurements.
  • Beyond the paper: the numerical $N=2$ robustness comparison suggests that higher-dimensional parallel strategies tolerate source noise better than qubit strategies with the same number of distant settings, a trend that could be checked systematically for $N=3$ with improved optimization methods.
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

2 major / 5 minor

Summary. The paper studies N-fold parallel repetitions of two routed Bell strategies, rBB84_N and rCHSH_N, and claims that when the near-source devices A and B0 maximally violate the N-product CHSH inequality, non-jointly-measurable (NJM) measurements in the distant device B1 can be certified for every detection efficiency η > 1/2^N. The threshold is argued to be tight because B1 has 2^N settings. The rBB84_N result is proven in Proposition 1 via a penalized BB84 inequality and a Popovici–Sebestyén bound on sums of tensor-product projectors. The rCHSH_N result is stated as Proposition 2 with a proof in Appendix B. The paper also presents numerical NPO results for N=2 showing improved robustness against source visibility, and a conditional analytic robustness statement in Proposition 3 based on an assumed robust self-testing bound. The central ideal-case result, if correct, would give an optimal exponential reduction of the critical detection efficiency for a DIQKD-relevant task.

Significance. If the claims are established, the paper represents a significant step for device-independent quantum cryptography: it constructs explicit routed Bell strategies for which the critical detection efficiency for certifying NJM measurements decreases as 2^{-N}, which is optimal given the number of settings of the distant device. The rBB84_N part is proven analytically with a clean argument, and the paper provides reproducible code and numerical evidence for the N=2 robustness comparison. The main caveats are that the analytic proof of the rCHSH_N result currently relies on a false bound in Appendix B, and the 'robust' claim in the abstract is not backed by an explicit robust self-testing function. These issues are local and likely repairable, but they need to be addressed before the full claims can be accepted.

major comments (2)
  1. Eq. (B17) claims that for any two distinct operators π' in the set {½(|0⟩⟨0|+|+⟩⟨+|), ½(|0⟩⟨0|+|−⟩⟨−|), ½(|1⟩⟨1|+|+⟩⟨+|), ½(|1⟩⟨1|+|−⟩⟨−|)} one has ||√π'√π''|| ≤ β = (2+√2+√(4√2−2))/8 ≈ 0.6658. This is false. Taking π' = ½(|0⟩⟨0|+|+⟩⟨+|) and π'' = ½(|0⟩⟨0|+|−⟩⟨−|), both of which are in the stated set, a direct calculation gives ||√π'√π''||² = λ_max(π'π'') = (2+√3)/8, hence ||√π'√π''|| ≈ 0.6830 > β. Consequently the elementwise bound Γ ≤ G_k in Eq. (B18) does not follow, and the subsequent bound ||Γ|| ≤ ||G_k||, which is the step that yields the q-range q ≥ α^{N−1}β, is not justified. As written, the analytic proof of Proposition 2 is incomplete; the exhaustive search for N ≤ 3 and the conjecture for N > 3 do not repair the proof for general N. The claimed threshold η* = 1/2^N for rCHSH_N may still be true with a corrected constant, but that corrected proof is not in the manuscript.
  2. The abstract states that the exponential decrease is achieved 'robustly', but Proposition 3 is conditional on an unquantified robust self-testing function f(N,ε) in Eq. (40). No explicit robust self-testing bound for the N-product CHSH inequality is provided, and the analogous robustness statement for rCHSH_N is explicitly deferred to future work. In addition, the proposition does not ensure that the chosen penalty q = 1/√2 + √δ remains in the admissible range [1/√2, 1] of Proposition 1, since δ is only bounded by an expression involving f(N,ε). As a result, the 'robustly' and 'arbitrary distances' claims in the abstract and conclusions are stronger than what the analytic results support. The robustness statement should be qualified, or the missing robust self-testing input should be supplied with explicit bounds.
minor comments (5)
  1. The sentence 'We assume that the detection efficiency is the same for all measurements in B1 and therefore p(b = ∅|y) = η' is inconsistent with η being the detection efficiency (click probability); it should read p(b = ∅|y) = 1 − η. The strategy value in Eq. (16) is consistent with the latter convention.
  2. The notation ||M_k|| is used for the maximal eigenvalue of M_k, whereas elsewhere in the paper ||·|| denotes the operator norm. Since M_k can have negative eigenvalues, these two quantities differ; please introduce a distinct symbol such as λ_max(M_k).
  3. There is a typographical error: 'f_k ≡ ... +−kq' should read 'f_k ≡ ... − kq'.
  4. The statement that an exhaustive search for N=1,2,3 yields β' and that the bound is conjectured for N>3 should be clearly separated from the analytic claim of Proposition 2; currently the text reads as if Proposition 2 itself relies on the conjecture, which is not the case.
  5. The normalization of the operators S_j in Appendix C should be reconciled with the conditional states in Eq. (14): in the ideal case the conditional states have trace one, whereas the S_j defined in Eq. (C8) appear to be sub-normalized. Please clarify this so that the constants in the threshold (41) can be verified.

Circularity Check

0 steps flagged · score 0.0 of 10

No circularity: the exponential threshold is derived from external self-testing bounds and independent operator-norm estimates, not from its own conclusion.

full rationale

The paper's central claim—that η* = 1/2^N certifies non-jointly-measurable measurements in B1 for rBB84_N and rCHSH_N, given perfect A and B0—is not obtained by fitting or by definition. The derivation starts from the externally established self-testing statement for the maximal violation of the N-product CHSH inequality (Eqs. (8)–(10), citing [27–29]) and uses two of its consequences: uniform Alice marginals (13) and the form of Bob's remotely prepared states (14). These are independent inputs that do not contain the target threshold. The JM upper bounds in Propositions 1 and 2 are derived by bounding the maximal eigenvalue of the operators C_b via the Popovici–Sebestyén bound and the elementwise matrix inequality of Appendix A; the strategy values (16) and (37) are computed directly from the proposed POVMs. Equating an independent upper bound with the strategy value yields η > 1/2^N by elementary algebra. The lower bound η* ≥ 1/2^N is quoted from the 2^N-settings counting argument in [17,26,30], an external result, not the paper's own conclusion. The use of the authors' earlier works [16,17] is architectural: they provide the routed-Bell and parallel-repetition frameworks but do not themselves contain the N-fold rBB84/rCHSH threshold. No parameter is fitted to the quantity being predicted; q is a free penalty variable and β is computed, not calibrated. The possible error in the numerical estimate (B17) flagged by a critical reader concerns the validity of an operator-norm bound and would affect correctness, not circularity: a false inequality cannot make the derivation equivalent to its inputs. Proposition 3 is explicitly conditional on an externally supplied robust self-testing function f(N,ε), and conditional claims are not circular. The paper is therefore self-contained against external benchmarks, and no circular step is exhibited.

Assumptions & free parameters 1 free parameters · 6 assumptions · 0 invented entities

No new physical entities are introduced; the rBB84_N and rCHSH_N strategies are protocols, not entities. The central claim rests on the perfect near-device assumption, the self-testing theorem, and the Norm bound.

free parameters (1)
  • q (penalty parameter) = q=1/sqrt(2) for rBB84_N; q in [alpha^(N-1)*beta, alpha^N] for rCHSH_N
    A mathematical penalty inserted into the inequalities (15) and (36); chosen so that the JM bounds (17) and (38) become tight. It is not fitted to data but is a hand-tuned parameter of the proof.
assumptions (6)
  • domain assumption Maximal violation of the N-product CHSH inequality self-tests N copies of the EPR state and the ideal measurements (cited to [27-29]).
    Used to obtain the marginal uniformity (13) and remote-state form (14), the backbone of Propositions 1 and 2. The paper does not prove this self-testing statement.
  • domain assumption Near-source devices A and B0 are perfect, i.e., achieve CHSH_N = alpha^N.
    Stated at the start of Section IV. Removes all loss and noise from the characterization step; without it the exact eta*=1/2^N proof does not go through.
  • domain assumption Detection efficiency eta is identical for all B1 measurements and failures produce a no-click outcome empty.
    Assumed before Eq. (16); needed for the simple strategy values (1-q)eta and (alpha^N-q)eta.
  • standard math For projective measurements, joint measurability is equivalent to commutativity of the POVM elements (used in the NPA robustness section).
    Used in Section V to reduce JM constraints to commutation relations, following [22].
  • standard math Popovici-Sebestyen norm bound for sums of positive semidefinite operators (Eq. 29; proof deferred to [36]).
    Central tool bounding the norm of the operators C_vecb in Propositions 1, 2, and 3.
  • domain assumption Existence of a robust self-testing statement of form (40) with some f(N,epsilon) tending to 0 (Proposition 3).
    The robustness result for N>2 is conditional on this existence; no concrete f(N,epsilon) for the N-product CHSH inequality is derived in the paper.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Extending quantum correlations to arbitrary distances via parallel repetition of routed Bell tests." pith.science (2026). https://pith.science/paper/VQRATIJ3

@misc{pith2026250417621,
  author       = {Pith},
  title        = {Pith review of: Extending quantum correlations to arbitrary distances via parallel repetition of routed Bell tests},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/VQRATIJ3}},
  note         = {Machine review of arXiv:2504.17621}
}
abstract

Applications such as Device-Independent Quantum Key Distribution (DIQKD) require loophole-free certification of long-distance quantum correlations. However, these distances remain severely constrained by detector inefficiencies and unavoidable transmission losses. To overcome this challenge, we consider parallel repetitions of the recently proposed routed Bell experiments, where transmissions from the source are actively directed either to a nearby or a distant measurement device. We analytically show that the threshold detection efficiency of the distant device--needed to certify non-jointly-measurable measurements, a prerequisite of secure DIQKD--decreases exponentially, optimally, and robustly, following $\eta^*=1/2^N$, with the number $N$ of parallel repetitions.

Figures

Figures reproduced from arXiv: 2504.17621 by the authors.

Figure 2
Figure 2. FIG. 2 [PITH_FULL_IMAGE:figures/full_fig_p002_2.png] view at source ↗
Figure 3
Figure 3. FIG. 3. Comparison of the upper bounds on the critical de [PITH_FULL_IMAGE:figures/full_fig_p007_3.png] view at source ↗

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

44 extracted references · 21 canonical work pages

  1. [38]

    B. S. Cirel’son, Letters in Mathematical Physics 4, 93 (1980)

    1980

  2. [1]

    J. S. Bell, Physics Physique Fizika 1, 195 (1964)

    1964

  3. [2]

    In the following, we show that||Mk||≤||M1||for all possi- ble settings, and therefore ||M1||= 1− 1√ 2 upper bounds the maximal eigenvalue of any possible C⃗b

    It is easy to compute the norm of 0-click and 1-click operators: ||M0||= 0 (26) ||M1||=||S1||− 1√ 2 = 1− 1√ 2 >||M0||, (27) 6 irrespective of y, since||Sl||= 1 as these are projections. In the following, we show that||Mk||≤||M1||for all possi- ble settings, and therefore ||M1||= 1− 1√ 2 upper bounds the maximal eigenvalue of any possible C⃗b. The norm of ...

    work page

  4. [3]

    (31) This implies that the matrix Γ is upper bounded element- wise by the matrix Gk =   1 1√ 2 1√ 2 ··· 1√ 21√ 2 1 1√ 2 ··· 1√ 2

    The rest of the norms are bounded by 1, and therefore in general we have Γl,l′≤ ( 1, if l =l′, 1√ 2, if l̸=l′. (31) This implies that the matrix Γ is upper bounded element- wise by the matrix Gk =   1 1√ 2 1√ 2 ··· 1√ 21√ 2 1 1√ 2 ··· 1√ 2 ... ... ... 1√ 2 1√ 2 1√ 2 ··· 1   (32) = 1− 1√ 2 I +   1√ 2 1√ 2 1√ 2 ··· 1√ 21√ 2 1√ 2 1√ 2 ··· 1...

    work page

  5. [4]

    central hub

    and has been deferred to Appendix C for brevity. We expect a similar robustness statement to hold for theq-penalized N-product CHSH inequality, but the de- tails are left for a future study. VI. CONCLUSIONS The combined effect of inevitable transmission losses and the limited efficiencies of detectors constitutes one of the most persistent challenges in t...

    work page 2022

  6. [5]

    Barrett, L

    J. Barrett, L. Hardy, and A. Kent, Phys. Rev. Lett. 95, 010503 (2005)

    2005

  7. [6]

    Brunner, D

    N. Brunner, D. Cavalcanti, S. Pironio, V. Scarani, and S. Wehner, Rev. Mod. Phys. 86, 419 (2014)

    2014

  8. [7]

    A. K. Ekert, Phys. Rev. Lett. 67, 661 (1991)

    1991

Show all 44 references
  1. [8]

    Mayers and A

    D. Mayers and A. Yao, in Proceedings 39th Annual Sym- posium on Foundations of Computer Science (IEEE, Los Alamitos, CA, 1998) p. 503

    1998

  2. [9]

    J. F. Clauser and M. A. Horne, Phys. Rev. D 10, 526 (1974)

    1974

  3. [10]

    Ac´ ın, N

    A. Ac´ ın, N. Brunner, N. Gisin, S. Massar, S. Pironio, and V. Scarani, Phys. Rev. Lett. 98, 230501 (2007)

    2007

  4. [11]

    Pironio, A

    S. Pironio, A. Ac´ ın, N. Brunner, N. Gisin, S. Massar, and V. Scarani, New J. Phys. 11, 045021 (2009)

    2009

  5. [12]

    P. M. Pearle, Phys. Rev. D 2, 1418 (1970)

    1970

  6. [14]

    P. H. Eberhard, Phys. Rev. A 47, R747 (1993)

    1993

  7. [15]

    Massar, Phys

    S. Massar, Phys. Rev. A 65, 032121 (2002)

    2002

  8. [16]

    Massar and S

    S. Massar and S. Pironio, Phys. Rev. A 68, 062109 (2003)

    2003

  9. [17]

    Chaturvedi, G

    A. Chaturvedi, G. Viola, and M. Paw lowski, npj Quan- tum Information 10, 7 (2024)

    2024

  10. [18]

    V´ ertesi, S

    T. V´ ertesi, S. Pironio, and N. Brunner, Phys. Rev. Lett. 104, 060401 (2010)

    2010

  11. [19]

    Z.-P. Xu, J. Steinberg, J. Singh, A. J. L´ opez-Tarrida, J. R. Portillo, and A. Cabello, Graph-theoretic approach to bell experiments with low detection efficiency (2022)

    2022

  12. [20]

    Miklin, A

    N. Miklin, A. Chaturvedi, M. Bourennane, M. Paw lowski, and A. Cabello, Phys. Rev. Lett. 129, 230403 (2022)

    2022

  13. [21]

    E. Y.-Z. Tan and R. Wolf, Phys. Rev. Lett. 133, 120803 (2024)

    2024

  14. [22]

    P. G. Kwiat, Journal of Modern Optics 44, 2173 (1997), https://www.tandfonline.com/doi/pdf/10.1080/09500349708231877

    1997 doi

  15. [23]

    Ara´ ujo, F

    M. Ara´ ujo, F. Hirsch, and M. T. Quintino, Quantum 4, 353 (2020)

    2020

  16. [24]

    C. C. W. Lim, C. Portmann, M. Tomamichel, R. Renner, and N. Gisin, Phys. Rev. X 3, 031006 (2013)

    2013

  17. [25]

    Gigena, E

    N. Gigena, E. Panwar, G. Scala, M. Ara´ ujo, M. Farkas, and A. Chaturvedi, Robust self-testing of bell inequali- ties tilted for maximal loophole-free nonlocality (2024), arXiv:2405.08743 [quant-ph]

    2024 arXiv

  18. [27]

    Masini, M

    M. Masini, M. Ioannou, N. Brunner, S. Pironio, and P. Sekatski, Quantum 8, 1574 (2024)

    2024

  19. [28]

    P. H. Eberhard and P. Rosselet, Found. Phys. 25, 91 (1995)

    1995

  20. [29]

    ˇSupi´ c, D

    I. ˇSupi´ c, D. Cavalcanti, and J. Bowles, Quantum5, 418 (2021)

    2021

  21. [30]

    Le Roy-Deloison, E

    T. Le Roy-Deloison, E. P. Lobo, J. Pauwels, and S. Piro- nio, PRX Quantum 6, 020311 (2025)

    2025

  22. [31]

    A. W. Coladangelo, Parallel self-testing of (tilted) EPR pairs via copies of (tilted) CHSH (2017), arXiv:1609.03687 [quant-ph]

    2017 arXiv

  23. [32]

    McKague, Quantum 1, 1 (2017)

    M. McKague, Quantum 1, 1 (2017)

    2017

  24. [33]

    Pironio, M

    S. Pironio, M. Navascu´ es, and A. Ac´ ın, SIAM Journal on Optimization 20, 2157 (2010), https://doi.org/10.1137/090760155

    2010 doi

  25. [34]

    E. P. Lobo, J. Pauwels, and S. Pironio, Quantum 8, 1332 (2024)

    2024

  26. [35]

    Navascu´ es, S

    M. Navascu´ es, S. Pironio, and A. Ac´ ın, Phys. Rev. Lett. 98, 010401 (2007)

    2007

  27. [36]

    not being tight, also leading to the lower bound on q [q ≥ αN−1β, for (38)] not being tight. An ex- haustive search over the (2N + 1) 2N operators{C⃗b}⃗b for 7 N = 1, 2, 3 yields a tighter lower boundq≥αN−1β′ (38), with β′ = 4− √ 2 4 ≈ 0.64644≤ β, which we conjecture to also h...

  28. [37]

    Navascu´ es, S

    M. Navascu´ es, S. Pironio, and A. Ac´ ın, New Journal of Physics 10, 073013 (2008)

    2008

  29. [39]

    McKague, T

    M. McKague, T. H. Yang, and V. Scarani, Journal of Physics A: Mathematical and Theoretical 45, 455304 (2012)

    2012

  30. [40]

    SEBESTY ´EN and D

    Z. SEBESTY ´EN and D. POPOVICI, Glasgow Mathe- matical Journal 47, 471–488 (2005)

    2005

  31. [41]

    ex- tending quantum correlations to arbitrary distances via parallel repetition of routed bell tests

    A. Chaturvedi, ParrelRepRoutedBell: Codes for “ex- tending quantum correlations to arbitrary distances via parallel repetition of routed bell tests”, https: //github.com/anubhav70747/ParrelRepRoutedBell (2025), gitHub repository, commit 6305f01, accessed 17 Apr 2025

    2025

  32. [42]

    Sekatski, J

    P. Sekatski, J. Pauwels, E. P. Lobo, S. Pironio, and N. Brunner, Certification of quantum correlations and diqkd at arbitrary distances through routed bell tests (2025), arXiv:2502.12241 [quant-ph]

    2025 arXiv

  33. [43]

    Wittek, ACM Transactions on Mathematical Software (TOMS) 41, 1 (2015)

    P. Wittek, ACM Transactions on Mathematical Software (TOMS) 41, 1 (2015)

    2015

  34. [44]

    L¨ ofberg, in Proceedings of the CACSD Conference (Taipei, Taiwan, 2004)

    J. L¨ ofberg, in Proceedings of the CACSD Conference (Taipei, Taiwan, 2004)

    2004

  35. [45]

    ApS, The MOSEK optimization toolbox for MATLAB manual

    M. ApS, The MOSEK optimization toolbox for MATLAB manual. Version 9.0. (2019)

    2019

  36. [46]

    A. E. Rastegin, Journal of Statistical Physics 148, 1040 (2012). 10 Appendix A: Element-wise inequality implies operator norm inequality Lemma 1. LetA,B∈Mn(C) such that 0≤Aij≤Bij for all i,j . Then ||A||≤||B||. Proof. By definition, ||A||= max x∈Cn ||x||=1 ||Ax||= max x∈Cn ||x...

    2012

Pith tools

Reviewed August 16, 2026 · model on record in the stance chip above.