Pith. sign in

REVIEW 2 major objections 3 minor 1 cited by

Time evolution under a fixed Hamiltonian can promote a state 1-design into a state k-design, and a mixed-field Ising chain numerically realizes this from Y-basis product states.

Reviewed by Pith at T0; open to challenge. T0 means a machine referee read the full paper against a public rubric. the ladder, T0–T4 →

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

T0 review · deepseek-v4-flash

2026-08-01 15:07 UTC pith:VSXSZK3N

load-bearing objection Solid GUE recursion, plausible but conditional local-Hamiltonian claim — worth refereeing, with the Porter-Thomas ansatz as the main issue to press. the 2 major comments →

arxiv 2607.18537 v1 pith:VSXSZK3N submitted 2026-07-20 quant-ph cond-mat.stat-mech

State k-designs from Hamiltonian evolution

Shengxian Hou , Zong-Yue Hou , Zhi-Cheng Yang This is my paper
classification quant-ph cond-mat.stat-mech
keywords state k-designframe potentialHamiltonian evolutionGUEPorter-Thomas statisticsmixed-field Ising modeltemporal ensemblefinite-time correction
verification ladder T0 review T1 audit T2 compute T3 formal T4 reserved

The pith

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

The paper asks when deterministic Hamiltonian evolution can make an ensemble of quantum states look Haar-random up to k-th moments. Its central result is a recursion relation for the GUE-averaged frame potential: after a long evolution time, the frame potential of the evolved ensemble is expressed directly through the frame potentials of the initial ensemble. If the initial states form a state 1-design, the evolved ensemble becomes a state k-design as the Hilbert-space dimension grows. The paper also gives numerical evidence that a fixed nonintegrable mixed-field Ising Hamiltonian generates approximate state k-designs up to k=5 from Y-basis product states, with error falling exponentially in system size, and proposes an M-step alternating quench protocol that suppresses finite-time corrections from O(1/T) to O(1/T^M). The same recursion explains recent unitary-design constructions.

Core claim

For Hamiltonians drawn from the GUE, and assuming the k-th no-resonance condition on the spectrum, the long-time GUE-averaged frame potential obeys E_H F_E^(k) = k!/D^k (1 + sum_{m=1}^k C(k,m) F_{E'}^{(m)}) plus subleading corrections. Because state frame potentials do not climb with order, an initial ensemble whose first frame potential is O(1/D) — in particular any state 1-design such as a complete orthonormal basis — is enough to make the evolved ensemble a state k-design in the thermodynamic limit. The same mechanism is shown to work numerically for a fixed local mixed-field Ising Hamiltonian (J,hx,hz)=(1,0.9045,0.809) when initial states are product bitstrings in the Y-basis: scaled fra

What carries the argument

The state frame potential F_E^(k) = E_{|psi>,|phi>~E} |<psi|phi>|^(2k), whose Haar minimum is k!/D^k, is the working object; an ensemble is a state k-design iff its frame potential equals that minimum. The driving identity is the GUE-averaged recursion in Theorem 1, which converts long-time evolution into a sum over frame potentials of the initial ensemble. For the local-Hamiltonian result, the Porter-Thomas statistics of squared eigenstate overlaps x_nm=|langle E_n|m>|^2 (with only E[x]=1/D and E[x^p]=Theta(D^-p) needed) turns Eq. (6) into the Haar value at leading order. The k-th no-resonance condition selects the long-time diagonal part of the time average; finite-T corrections are carrie

Load-bearing premise

The local-Hamiltonian claim rests on treating squared eigenstate overlaps of Y-basis product states as independent Porter-Thomas random variables — an assumption not derived from the Ising Hamiltonian and checked only numerically up to N=13 — plus the assumed k-th no-resonance of the fixed Ising spectrum; if either fails, the exponential 2^-N error bound collapses.

What would settle it

Exact-diagonalize the mixed-field Ising chain at N=14-16 and estimate the long-time frame potential from strata with |t-t'|>t0; if the relative error deltaF^(k) stops following 2^-N for k up to 5, or if a direct check of the k-th no-resonance condition finds a resonance among the eigenenergies, the claim that this fixed Hamiltonian generates approximate k-designs is falsified.

Watch this falsifier — get emailed when new claim-graph text bears on it.

If this is right

  • Any state 1-design of inputs — including a complete orthonormal basis — becomes a state k-design after long-time evolution under a GUE Hamiltonian, in the thermodynamic limit.
  • Two independent GUE Hamiltonian quenches suffice: one quench turns a fixed state into an ensemble with first-order frame potential O(1/D), and the second promotes it to a k-design.
  • For the mixed-field Ising chain, Y-basis product states produce approximate k-designs for k<=5 with error scaling as 2^-N, showing by exact diagonalization that local chaotic Hamiltonians can do the job.
  • The M-step alternating-quench protocol suppresses the finite-time error to O(1/T^M), so convergence can be reached with total evolution time linear in system size.
  • The unitary version of the recursion explains why three Hamiltonian quenches, or two quenches plus a random Pauli, are sufficient constructions for unitary k-designs.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • The role of the Y-basis suggests a testable design criterion for any local Hamiltonian: choose a product basis in which each state has zero first moment and reproduces the infinite-temperature second moment of H; other Clifford-rotated bases may work without extra fine-tuning.
  • If the Porter-Thomas independence assumption survives at larger N, design generation should be generic for chaotic Hamiltonians, not special to this Ising point; varying (hx,hz) while keeping nonintegrability would map the regime of validity.
  • The M-step protocol is effectively a pulse sequence; one could optimize the two Hamiltonians rather than fixing them, or probe the design via randomized-measurement protocols, opening a route to certification of designs from a single quench.

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

2 major / 3 minor

Summary. The paper studies ensembles E={e^{-iHt}|ψ0⟩ : t∼Unif[0,T], |ψ0⟩∼E′} and makes four main claims. For GUE Hamiltonians it derives, in the T→∞ limit, a recursion for the GUE-averaged state frame potential (Theorem 1): E_H F_E^(k) = k!/D^k (1 + Σ_{m=1}^k C(k,m) F_{E′}^{(m)}) + subleading, so any initial state 1-design yields an asymptotic state k-design. For a fixed mixed-field Ising Hamiltonian, it claims that Y-basis product states form approximate k-designs with error decaying as 2^{-N}, supported analytically by a Porter-Thomas ansatz and numerically up to N=13. It proves a no-go theorem (Theorem 2) that explains why Z- and X-bases fail, analyzes the finite-T correction as O(1/T), proposes an M-step quench protocol with O(1/T^M) suppression, and derives an analogous unitary recursion (Theorem 3) that unifies recent sequential-quench unitary-design constructions.

Significance. The GUE state and unitary recursions are clean, parameter-free results that, if correct, provide a unified mechanism for several recent constructions of state and unitary k-designs from Hamiltonian evolution. The no-go theorem and the finite-T GUE analysis are useful contributions. The local-Hamiltonian claim is the most striking and potentially significant part of the paper, but it rests on an unproven random-matrix ansatz and on numerics with limited system sizes. The manuscript is self-contained, with a detailed supplement and no fitted parameters; these are strengths. However, the analytical support for the fixed-Ising claim needs substantial strengthening before the paper can be accepted.

major comments (2)
  1. [Local Hamiltonians and SM Appendix 2, Eqs. (S10)-(S13)] The analytic claim F_E^(k)=k!/D^k for the fixed Ising model assumes the overlaps x_nm=|⟨E_n|m⟩|^2 are independent Porter-Thomas variables. This is inconsistent with unitarity: ∑_m x_nm=∑_n x_nm=1, so the variables cannot be independent. The key factorization E[∏ x_{n_a m} x_{n_a m′}]=D^{-2k} in Eq. (S13) is exactly that assumption. Even for a Haar-random H the exact two-point function is 1/[D(D+1)] for m≠m′, so the ansatz is an uncontrolled approximation; for a fixed H there is no averaging over H at all. Fig. 2(b) (N≤13, no error bars) is insufficient to establish that the exponential trend persists. This is load-bearing for the local-Hamiltonian result; GUE Theorems 1 and 3 are not affected. Please replace the ansatz by a derivation or a quantitative bound, and/or provide a direct numerical test of the factorization.
  2. [Temporal convergence and Fig. 3] The M-step O(1/T^M) suppression is not proven for the fixed local protocol; the rigorous finite-T derivation in SM Appendix 5 is for GUE-averaged Hamiltonians only. Moreover, combining the stated scaling with the Y-basis value F_{E′}^{(k)}=1/D gives an absolute error O(1/(D T^M)) after M steps; relative to F_Haar=k!/D^k this is O(D^{k-1}/T^M). With the proposed choice M=N and T=2k, this relative error behaves as O((2^{k-2}/k)^N), which does not decay for k≥4 (k=4 is constant, k=5 grows). Thus the claim that M=N and T=2k yields convergence in polynomial time is not supported by the paper's own scaling estimates. An explicit condition on T, or a revision of the claim, is needed.
minor comments (3)
  1. [Fig. 2 and Fig. 3] The numerical figures do not show statistical error bars. Since the End Matter describes Monte Carlo sampling with finite sample sizes (5×10^7 samples per stratum), error bars are feasible and should be included.
  2. [References] Some references have irregular titles or DOI-like identifiers (e.g., [16] "Nature is stingy", [18], [37] with nonstandard '10.1103/...' suffixes). Please verify all references and replace with the official journal/arXiv identifiers.
  3. [Eq. (2) and no-resonance condition] The k-th no-resonance condition is assumed for the fixed Ising Hamiltonian used in Fig. 2 but is not verified. Although generically expected for chaotic Hamiltonians, a numerical check for the specific parameters would strengthen the T→∞ limit.

Circularity Check

0 steps flagged

No significant circularity: derivations are parameter-free; the local-Hamiltonian result rests on an explicit external random-matrix ansatz, not on fitted inputs or load-bearing self-citation.

full rationale

Verdict: no significant circularity. The central GUE result (Theorem 1, Eq. 3) is derived from Eq. (2) via Weingarten/permutation calculus in SM Appendix 1; the only inputs are the stated k-th no-resonance condition and the definition of E', and no quantity is fitted to the data used to verify design formation. The local-Hamiltonian claim is explicitly conditional: the paper states that 'if the overlaps x_nm := |<E_n|m>|^2 obey the Porter-Thomas distribution and may be treated as independent random variables, then the leading contribution is precisely the Haar value, F_E^(k)=k!/D^k'. This is an external random-matrix assumption, not a restatement of the desired k-design property, and it is not calibrated against the Fig. 2 data; the numerics independently compare the final frame potential with the Haar value. The fact that the literal independence ansatz is incompatible with unitarity (row/column sums of |<E_n|m>|^2 are fixed) and that the paper itself notes spectral-edge states can deviate from Porter-Thomas statistics, while leaving open general criteria for local Hamiltonians, are rigor/correctness gaps rather than circular reductions. Theorem 2 is a self-contained energy-average argument; the M-step T^{-M} scaling is derived from the short-time analysis and separately verified numerically in Fig. 3; Theorem 3 is proved in SM Appendix 7 and is used to rederive the results of Refs. [19,20], not borrowed from them. No fitted parameter is relabeled as a prediction, and no load-bearing self-citation chain appears. Therefore the derivation chain is not circular.

Axiom & Free-Parameter Ledger

0 free parameters · 4 axioms · 0 invented entities

The central new mathematics rests only on standard quantum information definitions, no-resonance conditions, and Porter-Thomas statistics; no free parameters are fitted to the design-generation data, and no new entities are introduced.

axioms (4)
  • domain assumption k-th no-resonance condition on the eigenenergies of H
    Assumed for the GUE theorem (holds with probability 1) and implicitly for the local Ising numerics via Eq. (2). Entering the main text: 'We will assume the eigenenergies of H satisfy the k-th no-resonance condition...'
  • ad hoc to paper Eigenstate overlaps x_nm for Y-basis states behave as independent Porter-Thomas variables
    Used in SM Appendix 2 to show the local-Hamiltonian frame potential equals Haar at leading order; stated as a conditional 'if' in the main text near Eq. (6), not derived from the Hamiltonian.
  • standard math Weingarten calculus for GUE eigenbasis Haar averaging
    Standard random-matrix technique used in the proof of Theorem 1 (SM Appendix 1) and Theorem 3 (SM Appendix 7).
  • domain assumption Eigenstate thermalization/chaoticity of the mixed-field Ising spectrum
    Supports the claim that Y-basis states sample the chaotic bulk of the spectrum; cited Refs [27-30] and used in the main text discussion.

pith-pipeline@v1.3.0-alltime-deepseek · 26173 in / 22072 out tokens · 226013 ms · 2026-08-01T15:07:59.297785+00:00 · methodology

0 comments
read the original abstract

We study the generation of state $k$-designs from time evolution under a fixed Hamiltonian. Specifically, we consider the ensemble $\mathcal{E}=\left\{e^{-iHt}|\psi_0\rangle | \ t\sim \mathrm{Unif}[0,T],\, |\psi_0\rangle\sim \mathcal{E}'\right\}$, where the initial states are sampled from an ensemble $\mathcal{E}'$. For Hamiltonians drawn from the Gaussian unitary ensemble, we derive a simple relation between the frame potential of the evolved ensemble $\mathcal{E}$ and that of the initial ensemble $\mathcal{E}'$ in the large evolution time limit. This relation shows that $\mathcal{E}$ forms an exact state $k$-design in the thermodynamic limit as long as $\mathcal{E}'$ forms a state 1-design. Remarkably, we further show, both analytically and numerically, that time evolution under a simple nonintegrable mixed-field Ising Hamiltonian can generate approximate state $k$-designs with high precision, starting from product states in an appropriately chosen Pauli basis. We also analyze the finite-$T$ correction and find it scales as $O(1/T)$. To reduce the evolution time, we propose an $M$-step quench protocol that suppresses this correction to $O(1/T^M)$, which is also verified numerically. We then extend our analysis to unitary ensembles, deriving an analogous recursion relation for the unitary frame potential. Our results elucidate the mechanisms underlying recent proposals for generating unitary $k$-designs through sequential quantum quenches in a unified manner.

Figures

Figures reproduced from arXiv: 2607.18537 by Shengxian Hou, Zhi-Cheng Yang, Zong-Yue Hou.

Figure 1
Figure 1. Figure 1: FIG. 1. (a) The setup considered in this work, [PITH_FULL_IMAGE:figures/full_fig_p002_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: FIG. 2. (a) The scaled frame potential [PITH_FULL_IMAGE:figures/full_fig_p004_2.png] view at source ↗

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Forward citations

Cited by 1 Pith paper

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score.

  1. Unitary $k$-designs without Hamiltonian quenches

    quant-ph 2026-07 conditional novelty 7.0

    A single fixed chaotic Hamiltonian kicked twice by the same Pauli operator at randomly sampled times yields an approximate unitary k-design whose frame potential reaches the Haar value k! + O(1/d).

Reference graph

Works this paper leans on

42 extracted references · 3 canonical work pages · cited by 1 Pith paper

  1. [1]

    V. I. Arnold and A. Avez, Ergodic Problems of Classical Mechanics (W. A. Ben- jamin, New York, 1968)

    1968

  2. [2]

    I. P. Cornfeld, S. V. Fomin, and Y. G. Sinai, Ergodic Theory, Grundlehren der mathematischen Wis- senschaften, Vol. 245 (Springer, New York, 1982)

    1982

  3. [3]

    Haar, Annals of Mathematics34, 147 (1933)

    A. Haar, Annals of Mathematics34, 147 (1933)

    1933

  4. [4]

    J. M. Renes, R. Blume-Kohout, A. J. Scott, and C. M. Caves, Journal of Mathematical Physics45, 2171–2180 (2004)

    2004

  5. [5]

    A. J. Scott, Journal of Physics A: Mathematical and The- oretical41, 055308 (2008)

    2008

  6. [6]

    Ambainis and J

    A. Ambainis and J. Emerson, Quantum t-designs: t-wise independence in the quantum world (2007), arXiv:quant- ph/0701126

    arXiv 2007

  7. [7]

    Knill, D

    E. Knill, D. Leibfried, R. Reichle, J. Britton, R. B. Blakestad, J. D. Jost, C. Langer, R. Ozeri, S. Sei- delin, and D. J. Wineland, Physical Review A77, 10.1103/physreva.77.012307 (2008)

    doi:10.1103/physreva.77.012307 2008

  8. [8]

    Huang, R

    H.-Y. Huang, R. Kueng, and J. Preskill, Nature Physics 16, 1050–1057 (2020)

    2020

  9. [9]

    W. W. Ho and S. Choi, Physical Review Letters128, 10.1103/physrevlett.128.060601 (2022)

    doi:10.1103/physrevlett.128.060601 2022

  10. [10]

    Ippoliti and W

    M. Ippoliti and W. W. Ho, PRX Quantum4, 10.1103/prxquantum.4.030322 (2023)

    doi:10.1103/prxquantum.4.030322 2023

  11. [11]

    J. S. Cotler, D. K. Mark, H.-Y. Huang, F. Hern´ andez, J. Choi, A. L. Shaw, M. Endres, and S. Choi, PRX Quan- tum4, 010311 (2023)

    2023

  12. [12]

    J. Choi, A. L. Shaw, I. S. Madjarov, X. Xie, R. Finkel- stein, J. P. Covey, J. S. Cotler, D. K. Mark, H.-Y. Huang, A. Kale, H. Pichler, F. G. S. L. Brand˜ ao, S. Choi, and M. Endres, Nature613, 468–473 (2023)

    2023

  13. [13]

    W.-K. Mok, T. Haug, A. L. Shaw, M. Endres, and J. Preskill, Physical Review Letters134, 10.1103/phys- revlett.134.180403 (2025)

    doi:10.1103/phys- 2025

  14. [14]

    D. K. Mark, F. Surace, A. Elben, A. L. Shaw, J. Choi, G. Refael, M. Endres, and S. Choi, Phys. Rev. X14, 041051 (2024)

    2024

  15. [15]

    Ghosh, C

    S. Ghosh, C. M. Langlett, N. Hunter-Jones, and J. F. Rodriguez-Nieva, Physical Review B112, 10.1103/bwr6- vskn (2025)

    work page doi:10.1103/bwr6- 2025

  16. [16]

    W.-K. Mok, T. Haug, W. W. Ho, and J. Preskill, Nature is stingy: Universality of scrooge ensembles in quantum many-body systems (2026), arXiv:2601.00266 [quant-ph]

    Pith/arXiv arXiv 2026

  17. [17]

    Ghosh, N

    S. Ghosh, N. Hunter-Jones, and J. F. Rodriguez-Nieva, Randomization times under quantum chaotic hamilto- nian evolution (2025), arXiv:2512.25074 [cond-mat.stat- mech]

    arXiv 2025

  18. [18]

    Y.-N. Zhou, R. L¨ owenberg, and J. Sonner, Physical Re- view Letters136, 10.1103/rvwb-r9lv (2026)

    work page doi:10.1103/rvwb-r9lv 2026

  19. [19]

    Zhou, T.-G

    Y.-N. Zhou, T.-G. Zhou, and J. Sonner, Three Hamil- tonians are sufficient for unitaryk-design in temporal ensemble (2026), arXiv:2604.04205 [quant-ph]

    Pith/arXiv arXiv 2026

  20. [20]

    Sun and P

    N. Sun and P. Zhang, Unitary designs from two chaotic 6 Hamiltonians and a random Pauli operation (2026), arXiv:2604.10122 [quant-ph]

    Pith/arXiv arXiv 2026

  21. [21]

    L. Cui, T. Schuster, L. Mao, H.-Y. Huang, and F. Brand˜ ao, Random unitaries from hamiltonian dynam- ics (2025), arXiv:2510.08434 [quant-ph]

    arXiv 2025

  22. [22]

    Riddell, N

    J. Riddell, N. J. Pagliaroli, and ´Alvaro M. Alhambra, SciPost Phys.15, 165 (2023)

    2023

  23. [23]

    Huang and A

    Y. Huang and A. W. Harrow, Instability of localization in translation-invariant systems (2020), arXiv:1907.13392 [cond-mat.dis-nn]

    Pith/arXiv arXiv 2020

  24. [24]

    Nakata, M

    Y. Nakata, M. Koashi, and M. Murao, New Journal of Physics16, 053043 (2014)

    2014

  25. [25]

    See Supplemental Material for proofs of the theorems quoted in the main text, Porter-Thomas analysis for the product basis initial ensembles, derivation of the finite-T correction, and stabilizer state initial ensemble

  26. [26]

    See SM for more details

    In fact, we only require that the moments satisfy E[xnm] = 1/D,E[x p nm] = Θ(D −p), and that different xnm can be treated as independent variables. See SM for more details

  27. [27]

    Srednicki, Phys

    M. Srednicki, Phys. Rev. E50, 888 (1994)

    1994

  28. [28]

    Rigol, V

    M. Rigol, V. Dunjko, and M. Olshanii, Nature452, 854 (2008)

    2008

  29. [29]

    D’Alessio, Y

    L. D’Alessio, Y. Kafri, A. Polkovnikov, and M. Rigol, Adv. Phys.65, 239 (2016)

    2016

  30. [30]

    Y. Y. Atas and E. Bogomolny, J. Phys. A: Math. Theor. 50, 385102 (2017)

    2017

  31. [31]

    A. A. Mele, Quantum8, 1340 (2024), arXiv:2307.08956 [quant-ph]

    Pith/arXiv arXiv 2024

  32. [32]

    Kim and D

    H. Kim and D. A. Huse, Phys. Rev. Lett.111, 127205 (2013)

    2013

  33. [33]

    M. C. Ba˜ nuls, J. I. Cirac, and M. B. Hastings, Phys. Rev. Lett.106, 050405 (2011)

    2011

  34. [34]

    E. J. Torres-Herrera, M. T´ avora, and L. F. Santos, Brazil- ian Journal of Physics46, 239–247 (2015)

    2015

  35. [35]

    Cotler, N

    J. Cotler, N. Hunter-Jones, J. Liu, and B. Yoshida, Journal of High Energy Physics2017, 10.1007/jhep11(2017)048 (2017)

    doi:10.1007/jhep11(2017)048 2017

  36. [36]

    Y. V. Fyodorov, Introduction to the random matrix theory: Gaussian unitary ensemble and beyond (2010), arXiv:math-ph/0412017 [math-ph]

    Pith/arXiv arXiv 2010

  37. [37]

    Zhang, S

    Y. Zhang, S. Vijay, Y. Gu, and Y. Bao, PRX Quantum 7, 10.1103/myrb-nyhf (2026)

    work page doi:10.1103/myrb-nyhf 2026

  38. [38]

    Gross, S

    D. Gross, S. Nezami, and M. Walter, Communications in Mathematical Physics385, 1325–1393 (2021)

    2021

  39. [39]

    Bittel, J

    L. Bittel, J. Eisert, L. Leone, A. A. Mele, and S. F. E. Oliviero, A complete theory of the clifford commutant (2025), arXiv:2504.12263 [quant-ph]

    Pith/arXiv arXiv 2025

  40. [40]

    kY a=1 x2 nam # + k! D2 X m̸=m′ X ⃗ n 1 Ω(⃗ n)E

    R. O’Donnell, Analysis of boolean functions (2021), arXiv:2105.10386 [cs.DM]. End Matter F rame potential andk-designs Here we summarize the relation between the frame po- tential and the additive error of approximate designs. For a state ensembleE, itsk-th moment operator is defined as ρ(k) E := E |ψ⟩∼E h (|ψ⟩⟨ψ|) ⊗k i .(A1) Eforms a statek-design if and...

    Pith/arXiv arXiv 2021

  41. [41]

    LetP={B 1,· · ·, Bl}be a partition withl=|P|blocks

    Analysis ofG P (T) We first analyze the behavior of the functiong P (τ). LetP={B 1,· · ·, Bl}be a partition withl=|P|blocks. By definition, gP (τ) = Z DEexp [iτ(q B1 E1 +· · ·+qBl El)].(S32) We adopt the GUE convention used in Ref. [35]. For aD×DGUE matrix, its diagonal components are independent real Gaussian random variablesN(0,1/D) and its off-diagonal...

  42. [42]

    MP =M(⃗ n, ⃗ n′) = E |ψ⟩,|ϕ⟩∼E ′ X π,σ∈S 2k Wg(π−1σ, D)tr[VD(σ−1)(|ϕ⟩ ⟨ϕ|⊗k ⊗ |ψ⟩ ⟨ψ|⊗k)]⟨⃗ n, ⃗ n′|V D(π)|⃗ n′, ⃗ n⟩,(S55) Here we introduce some notation

    Analysis ofM P We first recall the definition ofM P , which is equal toM(⃗ n, ⃗ n′) for any⃗ x= (⃗ n, ⃗ n′) withP(⃗ x) =P. MP =M(⃗ n, ⃗ n′) = E |ψ⟩,|ϕ⟩∼E ′ X π,σ∈S 2k Wg(π−1σ, D)tr[VD(σ−1)(|ϕ⟩ ⟨ϕ|⊗k ⊗ |ψ⟩ ⟨ψ|⊗k)]⟨⃗ n, ⃗ n′|V D(π)|⃗ n′, ⃗ n⟩,(S55) Here we introduce some notation. LetP={B 1,· · ·, Bl}be a partition withl=|P|blocks. Pick a specific element i...