Pith. sign in

REVIEW 1 major objections 5 minor 40 references

Finite-N gauge-invariant operators of d matrices inevitably develop a non-free 'secondary' sector by degree about 2 log_d N, far before the first trace identity at degree N+1.

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 03:53 UTC pith:P4GOVFJX

load-bearing objection The algebraic core—single-trace hsop plus log-N onset of secondary invariants—is new and sound; the fast-scrambling claim is a clearly labeled speculative bridge. the 1 major comments →

arxiv 2607.23025 v1 pith:P4GOVFJX submitted 2026-07-25 hep-th

Overcrowding and the Finite-N Hilbert Space

Robert de Mello Koch , Anik Rudra , Augustine Larweh Mahu This is my paper
classification hep-th
keywords Hironaka decompositionprimary and secondary invariantsfinite-N matrix modelstrace identitiessingle-trace operatorsfast scramblingnecklace countinggauge-invariant operators
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 proves that in the algebra of gauge-invariant operators built from d Hermitian N×N matrices, the usual large-N picture of freely generated single-trace operators must break down by a degree that grows only as 2 log_d N + log_d log_d N. Specifically, one can always choose the independent 'primary' coordinates to be single traces, but no matter how they are chosen, at least one single-trace direction is left over at or before that degree and becomes a 'secondary' invariant — an intrinsically finite-N piece of data. The mechanism is pure counting: the number of independent short traces grows exponentially with length, while only 1+(d-1)N^2 of them can serve as free coordinates. This happens parametrically before the first trace identity at degree N+1, so it is a capacity effect, not a linear-dependence effect. The paper argues that the same counting, combined with ballistic growth of operator word length, gives the log-N scrambling time of fast scramblers, and demonstrates in low-rank examples that secondary invariants can label semiclassical sectors connected by instantons.

Core claim

The central claim is Theorem 3.1 plus Corollary 4.2. For every N≥1 and d≥2, the invariant ring R_{N,d} admits a homogeneous system of parameters consisting entirely of single-trace operators. Fix any such all-single-trace system P; if the first degree at which a single-trace direction is not used as a primary is m* ≤ N, then the quotient of the ring by the primaries has its first nonzero homogeneous piece at degree m*, canonically isomorphic to the unused part of the single-trace space (Q_P)_{m*} ≅ T_{m*}/U_{m*}. Since the number of single traces of length ≤ L, counted as necklaces, exceeds h = 1+(d-1)N^2 already at L = 2 log_d N + log_d log_d N + O_d(1), every such decomposition has δ(P) ≤

What carries the argument

The machinery is the Hironaka decomposition of the invariant ring R_{N,d} = ⊕ η_α P over the polynomial ring P generated by h primary invariants, together with necklace counting of single traces. The critical identity is Theorem 4.1: in the stable range m* ≤ N, the first secondary space equals the quotient T_{m*}/U_{m*}, so that the first 'missing' single-trace direction is exactly the first secondary. Burnside's lemma counts cyclic words of length m, giving dim T_m = a_d(m), and the cumulative count A_d(L) overtakes h at the logarithmically small scale L_{N,d}. The proof of Theorem 3.1 uses a common-degree lemma to cut down all invariant branches with single-trace linear combinations, ensur

Load-bearing premise

The load-bearing premise for the scrambling claim is dynamical, not algebraic: the paper assumes the trace-word length of a chaotic operator grows ballistically (L(t) ≈ v_op t) and that chaotic evolution explores single-trace directions evenly, so the overcrowding count equals the scrambling count; without these, the O(log N) scale is purely kinematical.

What would settle it

A concrete check: simulate a chaotic multi-matrix quantum mechanics at finite N (e.g., a commutator-squared potential) and measure the growth of the characteristic trace-word length of a precursor operator. If L(t) grows sublinearly or saturates below L_{N,d} ~ 2 log_d N on the scrambling timescale, the identification of overcrowding with scrambling is refuted; conversely, a measurement of t_* ~ (2 log N)/λ_op would support it. For the algebraic bound itself, an explicit all-single-trace hsop with δ(P) > L_{N,d} at any N,d would refute Corollary 4.2.

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

If this is right

  • The exact finite-N invariant algebra of d matrices is never freely generated by single traces beyond word length ~2 log_d N; a secondary sector is unavoidable inside that logarithmic window.
  • In the gauged matrix harmonic oscillator, the singlet Hilbert space decomposes into primary towers plus a finite secondary module; the first secondary appears at excitation number O(log N), reorganizing the finite-N spectrum long before the giant-graviton scale N or the scale N log N where factorization fails.
  • If operator word length grows ballistically with time, the overcrowding scale equals the fast-scrambling time t_* ~ (2 log N + log log N)/λ_op, giving a counting-based, algebraic explanation of scrambling in matrix models.
  • The all-single-trace property of primaries preserves the holographic dictionary: the free, perturbative coordinates remain single-particle duals, while secondary invariants carry the intrinsically finite-N, non-perturbative information.
  • Low-rank (N=2) examples show secondary invariants can distinguish configurations with identical primary data, label distinct semiclassical sectors, and mediate transitions through finite-action instantons with exponential suppression.

Where Pith is reading between the lines

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

  • If the overcrowding bound is tight or near-tight for optimized primary choices, then the logarithmic onset is a universal kinematic feature of every multi-matrix invariant ring, independent of dynamics; any matrix model with these symmetries inherits it (the scrambling connection still requires dynamics).
  • Because the secondary module has e^{cN^2} elements while the first secondary appears at O(log N), the finite fiber over primary space grows rapidly with degree and N; this suggests a hierarchical structure of non-perturbative sectors worth mapping explicitly for small N,d.
  • One testable extension: compute the optimized onset Δ_{N,d} (the latest first secondary among all single-trace hsops) for N=2,3,4 at d=2,3 to see how close it sits to L_{N,d}; the paper leaves a quantitative lower bound open.
  • The toy and matrix-model examples suggest that potentials built only from low-degree primaries will naturally replicate their minima over the finite secondary fiber; a systematic search over primary-fixing potentials could reveal whether orientation and Gram-sheet instantons persist at larger N.

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

1 major / 5 minor

Summary. The paper studies the finite-N invariant ring R_{N,d} of d Hermitian N×N matrices under U(N) conjugation. Its main algebraic results are: (i) Theorem 3.1, that an all-single-trace homogeneous system of parameters (hsop) always exists; (ii) Theorem 4.1, that in the stable range m* ≤ N the first single-trace direction omitted from such an hsop is precisely the first nontrivial secondary invariant; and (iii) Corollary 4.2, that for every all-single-trace hsop the first secondary appears by degree δ(P) ≤ L_{N,d} = 2 log_d N + log_d log_d N + O_d(1), parametrically before the first trace identity at degree N+1. The mechanism is overcrowding: exponentially many short single traces compete for only h = 1+(d-1)N^2 primary slots. The paper then proposes that this logarithmic scale is the algebraic counterpart of fast scrambling, and illustrates the secondary-sector interpretation with a Z_2 toy model and with N=2 three- and four-matrix examples in which secondary invariants label distinct semiclassical sectors connected by explicit instantons.

Significance. If correct, Corollary 4.2 is a clean and nontrivial statement about the finite-N invariant algebra: a non-free secondary sector is unavoidable at degree O(log N), well below any trace identity. The proof is self-contained and parameter-free, and the key computations (necklace counts, Hilbert series, Gram determinants, Hironaka decompositions of the N=2 examples) are explicit and reproducible. The all-single-trace hsop theorem is also a useful structural result for collective field theory, since it reconciles the primary-coordinate picture with the holographic single-trace dictionary. The fast-scrambling proposal is speculative and is explicitly labelled by the authors as a dynamical assumption rather than a consequence of invariant theory; this is the weakest part of the physical narrative but it does not bear on the central algebraic claim. Overall, the paper is significant and publishable, with the caveat that the scrambling language in the abstract should be tempered.

major comments (1)
  1. [§5 (Eqs. 5.4–5.13)] The identification of overcrowding with fast scrambling is conditional on two extra assumptions: ballistic growth of trace-word length, L(t) ≃ v_op t, and 'democratic' exploration of the available single-trace directions. Neither is derived from the Hamiltonian, and n_spread(t) in Eq. (5.7) is not defined operationally. The authors themselves state that Eq. (5.4) is a dynamical assumption and that the algebraic result does not prove scrambling. I agree with that assessment. My request is therefore presentational: the abstract and introduction should make explicit that the scrambling connection is a conjecture based on an unproved dynamical premise, not a corollary of the invariant-theory result. This does not undermine Corollary 4.2.
minor comments (5)
  1. [§3.2 / Appendix A.2] In the proof of the common-degree lemma, the argument uses eigenvalues and phases of the complexified representative matrices. This is legitimate, but the transition from Hermitian configurations to arbitrary complex tuples should be flagged more prominently at the start of the construction, since the physical reader may otherwise wonder why the dominant-eigenvalue phase argument applies to the original Hermitian problem.
  2. [§4.3] The hypersurface example is clear and useful. It would help to state explicitly that the degrees of the secondaries in the non-optimized decomposition are 1, 2, and 3, and that this demonstrates the basis-dependence of δ(P) without changing the ring. This is already implicit, but a one-line degree listing would improve readability.
  3. [§8.4] The 'Stability' test is asserted with the statement that the Hessian was checked, but no Hessian or eigenvalue computation is shown. Since the instanton claims rely on the four points being isolated local minima, please include at least the leading Hessian eigenvalues or a brief argument that the couplings ensure positive definiteness.
  4. [§5] The conversion from overcrowding to scrambling uses λ_op = v_op log d. The quantity v_op is introduced as a model-dependent ballistic velocity, but its normalization and dependence on the Hamiltonian's degree r are not specified. A short comment on the expected parametric size of v_op in a concrete matrix model would help the reader assess whether the O(1) constants in Eq. (5.13) are meaningful.
  5. [§8.1] After Eq. (8.10), the degree list of the secondary basis is 0, 2, 3, 3, 3, 3, 4, 6. It would be convenient to record this explicitly next to the η-notation, since it is used immediately to match the numerator of the Hilbert series in Eq. (8.4).

Circularity Check

0 steps flagged

No significant circularity: Cor. 4.2 follows from external trace-identity results plus a self-contained counting argument; self-citations are not load-bearing.

full rationale

The central derivation is self-contained and non-circular. The number of primary slots h=1+(d-1)N^2 is computed from orbit-space dimension (Eq. 2.6). Theorem 3.1 is proved in-paper via the common-zero-locus criterion (Appendix A.1) and the branch-cutting/common-degree Lemma 3.2, using only the standard fact that traces generate the invariant ring; it does not assume the conclusion. Theorem 4.1 identifies the first missing single-trace direction with the first secondary; its nontrivial input is the stable-range fact that no trace identity occurs before degree N+1, cited to Procesi/Razmyslov rather than to the authors. Corollary 4.2 is then a pigeonhole/counting argument: the necklace estimate A_d(L)~d^{L+1}/((d-1)L) (Eq. 4.16) is equated with h to obtain L_N,d=2 log_d N+log_d log_d N+O_d(1), and A_d(L)>h forces a missing primary direction, which Theorem 4.1 converts into a secondary of degree at most L_N,d. There are no fitted parameters passed off as predictions. The fast-scrambling discussion is explicitly conditional: the paper states 'Equation (5.4) is therefore a dynamical assumption rather than a consequence of invariant theory alone' and 'the algebraic result by itself does not prove scrambling'; an openly labeled assumption is not a circular derivation. Self-citations [13-16] supply context, the motivational e^{cN^2} secondary count, and illustrative low-rank ring descriptions, but none of these is used to establish Corollary 4.2. Hence no circular step is present.

Axiom & Free-Parameter Ledger

2 free parameters · 6 axioms · 0 invented entities

The central bound consumes only: (i) the orbit-space dimension h = 1+(d-1)N^2 (generic stabilizer argument, Eq. 2.6); (ii) the necklace count (Burnside, Eq. 4.9); (iii) stable-range independence of trace monomials (Procesi/Razmyslov); plus the Cohen-Macaulay theorem to license the free module. No datum is fitted; lambda_op and v_op are placeholders for the dynamics assumptions of Section 5, not fitted numbers. The hsop criterion of Appendix A.1 is standard but proved only by sketch in the paper. The e^{cN^2} secondary count is cited from the authors' earlier work [15] and is motivational only. No new physical entities are introduced.

free parameters (2)
  • lambda_op (operator growth exponent)
    Defined as lambda_op = v_op log d in Eq. (5.5) and used to convert the overcrowding scale into the scrambling-time formula (5.13). It encodes the untested ballistic-growth and democratic-exploration assumptions; no value is specified, making the 'prediction' t* = (2 log N + log log N)/lambda_op parametric in an unknown dynamical quantity.
  • v_op (ballistic front velocity)
    Introduced in Eq. (5.4) as the velocity of trace-word length growth. Not fitted, but the entire overcrowding-scrambling identification runs through it; the algebraic results do not depend on it.
axioms (6)
  • standard math Hochster-Roberts: the invariant ring of a reductive group acting on a regular ring is Cohen-Macaulay, so the module over primaries is free
    Licenses the Hironaka decomposition (2.8) used throughout; cited to [24].
  • standard math The invariant ring is generated by single traces, and the first trace identity has degree N+1, so trace monomials are linearly independent in the stable range
    Load-bearing input for Theorem 4.1 (stable-range independence of T_{m*}/U_{m*}); cited to Procesi, Razmyslov, Drensky-Formanek [21-23].
  • domain assumption Imposing Hermiticity does not change the algebraic invariant problem: R_{N,d} is isomorphic to the GL_N(C) invariants of d complex matrices (Section 2)
    Bridges the physical Hermitian model to the complexified algebraic geometry used in the proof of Theorem 3.1 and the common-degree lemma.
  • domain assumption A generic tuple of d Hermitian matrices has central-U(1) stabilizer, giving orbit-space dimension h = 1 + (d-1)N^2 (Eq. 2.6)
    Sets the number of primary slots; the overcrowding inequality compares the necklace count to this h.
  • domain assumption The singlet Fock-space map f -> f(A^dag)|0> is injective and covers all singlet states (Eqs. 2.13-2.14)
    Transfers the ring decomposition to the Hilbert-space decomposition (2.16); standard for bosonic oscillators.
  • ad hoc to paper Ballistic growth of trace-word length and democratic exploration of single-trace directions (Eq. 5.4 and Section 5)
    The bridging assumption of the scrambling claim, explicitly flagged by the authors as a dynamical assumption that should be tested; without it the O(log N) scale is purely kinematical.

pith-pipeline@v1.3.0-alltime-deepseek · 26093 in / 45847 out tokens · 394698 ms · 2026-08-01T03:53:47.512189+00:00 · methodology

0 comments
read the original abstract

Finite-$N$ trace relations reorganize the Hilbert space of gauge-invariant operators beyond the freely generated large-$N$ description. We study this structure using the Hironaka decomposition of the invariant ring of $d$ Hermitian $N\times N$ matrices. We first prove that the primary invariants may always be chosen to be homogeneous single-trace operators. We then show that, for any such choice, a nontrivial secondary invariant must appear by degree $L_{N,d}=2\log_d N+\log_d\log_d N+\mathcal{O}_d(1)$, which is parametrically below the first universal trace identity at degree $N+1$. This is a global overcrowding effect: exponentially many independent short single traces compete for only $1+(d-1)N^2$ algebraically independent coordinates. The overcrowding scale matches the fastest scrambling times expected for fast scramblers. We argue that this agreement of scales is not accidental: overcrowding provides a microscopic algebraic picture of scrambling in matrix models. Low-rank examples show that secondary invariants can distinguish configurations with identical primary data and, for suitable dynamics, label semiclassical sectors connected by instantons. These results identify the Hironaka decomposition as a natural framework for organizing perturbative and intrinsically finite-$N$ information in collective descriptions of gauge theories.

Figures

Figures reproduced from arXiv: 2607.23025 by Anik Rudra, Augustine Larweh Mahu, Robert de Mello Koch.

Figure 1
Figure 1. Figure 1: A four well potential: A particle can be trapped in any given well. Perturbing it slightly causes the particle to oscillate about the bottom of the well and is described by using perturbation theory. Transitions between wells can be accomplished by quantum tunnelling. These effects are typically of order e −Sinst/ℏ with Sinst the instanton action. 7.2 Invariant Ring Following [16] write the three Hermitian… view at source ↗
Figure 2
Figure 2. Figure 2: Secondary Sheets over the Primary Space The primary data generally do not fix the secondary invariants, so the invariant configuration space forms a finite collection of sheets over the primary-coordinate space, distinguished by the secondary values. If the potential depends only on the primaries and has a minimum at a given primary point, that minimum is repli￾cated on each dynamically accessible sheet. S… view at source ↗

discussion (0)

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

Reference graph

Works this paper leans on

40 extracted references · 28 linked inside Pith

  1. [1]

    The largeNlimit of superconformal field theories and supergravity,

    J. M. Maldacena, “The largeNlimit of superconformal field theories and supergravity,” Adv. Theor. Math. Phys.2(1998) 231–252 [hep-th/9711200]

    Pith/arXiv arXiv 1998

  2. [2]

    Anti-de Sitter space and holography,

    E. Witten, “Anti-de Sitter space and holography,” Adv. Theor. Math. Phys.2(1998) 253–291 [hep-th/9802150]

    Pith/arXiv arXiv 1998

  3. [3]

    Gauge theory correlators from non-critical string theory,

    S. S. Gubser, I. R. Klebanov and A. M. Polyakov, “Gauge theory correlators from non-critical string theory,” Phys. Lett. B428(1998) 105–114 [hep-th/9802109]

    Pith/arXiv arXiv 1998

  4. [4]

    The Quantum Collective Field Method and Its Application to the Planar Limit,

    A. Jevicki and B. Sakita, “The Quantum Collective Field Method and Its Application to the Planar Limit,” Nucl. Phys. B165(1980), 511

    1980

  5. [5]

    Collective Field Approach to the LargeNLimit: Euclidean Field Theories,

    A. Jevicki and B. Sakita, “Collective Field Approach to the LargeNLimit: Euclidean Field Theories,” Nucl. Phys. B185(1981), 89-100

    1981

  6. [6]

    String Field Theory and Physical Interpretation ofD= 1 Strings,

    S. R. Das and A. Jevicki, “String Field Theory and Physical Interpretation ofD= 1 Strings,” Mod. Phys. Lett. A5(1990), 1639-1650

    1990

  7. [7]

    Large N collective fields and holography,

    S. R. Das and A. Jevicki, “Large N collective fields and holography,” Phys. Rev. D68 (2003), 044011 [arXiv:hep-th/0304093 [hep-th]]

    Pith/arXiv arXiv 2003

  8. [8]

    Gravitational dynamics from collective field theory,

    R. de Mello Koch, “Gravitational dynamics from collective field theory,” JHEP10(2023), 151 [arXiv:2309.11116 [hep-th]]

    Pith/arXiv arXiv 2023

  9. [9]

    AdS 4/CF T3 Construction from Collective Fields,

    R. de Mello Koch, A. Jevicki, K. Jin and J. P. Rodrigues, “AdS 4/CF T3 Construction from Collective Fields,” Phys. Rev. D83(2011), 025006 [arXiv:1008.0633 [hep-th]]

    Pith/arXiv arXiv 2011

  10. [10]

    AdS Maps and Diagrams of Bi-local Holography,

    R. de Mello Koch, A. Jevicki, K. Suzuki and J. Yoon, “AdS Maps and Diagrams of Bi-local Holography,” JHEP03(2019), 133 [arXiv:1810.02332 [hep-th]]

    Pith/arXiv arXiv 2019

  11. [11]

    Consistent equation for interacting gauge fields of all spins in (3+1)-dimensions,

    M. A. Vasiliev, “Consistent equation for interacting gauge fields of all spins in (3+1)-dimensions,” Phys. Lett. B243(1990), 378-382

    1990

  12. [12]

    Higher spin gauge theories in four-dimensions, three-dimensions, and two-dimensions,

    M. A. Vasiliev, “Higher spin gauge theories in four-dimensions, three-dimensions, and two-dimensions,” Int. J. Mod. Phys. D5(1996), 763-797 [arXiv:hep-th/9611024 [hep-th]]. – 38 –

    Pith/arXiv arXiv 1996

  13. [13]

    Structure of loop space at finiteN,

    R. de Mello Koch and A. Jevicki, “Structure of loop space at finiteN,” JHEP06(2025) 011 [arXiv:2503.20097 [hep-th]]

    Pith/arXiv arXiv 2025

  14. [14]

    Hilbert space of finiteNmulti-matrix models,

    R. de Mello Koch and A. Jevicki, “Hilbert space of finiteNmulti-matrix models,” JHEP11 (2025) 145 [arXiv:2508.11986 [hep-th]]

    Pith/arXiv arXiv 2025

  15. [15]

    From symmetry to structure: Gauge-invariant operators in multi-matrix quantum mechanics,

    R. de Mello Koch, M. Kim and H. J. R. Van Zyl, “From symmetry to structure: Gauge-invariant operators in multi-matrix quantum mechanics,” [arXiv:2507.01219 [hep-th]]

    arXiv

  16. [16]

    Secondary invariants and non-perturbative states,

    R. de Mello Koch and J. P. Rodrigues, “Secondary invariants and non-perturbative states,” [arXiv:2604.15600 [hep-th]]

    Pith/arXiv arXiv

  17. [17]

    Bosonic fortuity in vector models,

    R. de Mello Koch, A. Ghosh and H. J. R. Van Zyl, “Bosonic fortuity in vector models,” JHEP06(2025), 246 [arXiv:2504.14181 [hep-th]]

    Pith/arXiv arXiv 2025

  18. [18]

    FiniteNHilbert Spaces of Bilocal Holography,

    R. de Mello Koch, A. Jevicki and J. Yoon, “FiniteNHilbert Spaces of Bilocal Holography,” [arXiv:2602.20788 [hep-th]]

    arXiv

  19. [19]

    Finite-NOperator Algebras and the Hilbert Space of Bilocal Holography,

    B. Ahn, R. de Mello Koch, V. Garg, A. Jevicki and J. Yoon, “Finite-NOperator Algebras and the Hilbert Space of Bilocal Holography,” [arXiv:2607.04143 [hep-th]]

    Pith/arXiv arXiv

  20. [20]

    Nonperturbative collective field theory,

    A. Jevicki, “Nonperturbative collective field theory,” Nucl. Phys. B376(1992), 75-98

    1992

  21. [21]

    Procesi,The invariant theory ofn×nmatrices, Adv

    C. Procesi,The invariant theory ofn×nmatrices, Adv. Math.19(1976) 306–381

    1976

  22. [22]

    Yu. P. Razmyslov,Trace identities of full matrix algebras over a field of characteristic zero, Math. USSR-Izv.8(1974) 727–760

    1974

  23. [23]

    Drensky and E

    V. Drensky and E. Formanek,Polynomial Identity Rings, Advanced Courses in Mathematics CRM Barcelona, Birkh¨ auser (2004)

    2004

  24. [24]

    Hochster and J

    M. Hochster and J. L. Roberts,Rings of invariants of reductive groups acting on regular rings are Cohen–Macaulay, Adv. Math.13(1974) 115–175

    1974

  25. [25]

    Fast Scramblers,

    Y. Sekino and L. Susskind, “Fast Scramblers,” JHEP10(2008), 065 [arXiv:0808.2096 [hep-th]]

    Pith/arXiv arXiv 2008

  26. [26]

    Exact Multi-Matrix Correlators,

    R. Bhattacharyya, S. Collins and R. de Mello Koch, “Exact Multi-Matrix Correlators,” JHEP03(2008) 044 [arXiv:0801.2061 [hep-th]]

    Pith/arXiv arXiv 2008

  27. [27]

    Exact Multi-Restricted Schur Polynomial Correlators,

    R. Bhattacharyya, R. de Mello Koch and M. Stephanou, “Exact Multi-Restricted Schur Polynomial Correlators,” JHEP06(2008) 101 [arXiv:0805.3025 [hep-th]]

    Pith/arXiv arXiv 2008

  28. [28]

    Restricted Schur Polynomials for Fermions and integrability in thesu(2|3) sector,

    R. de Mello Koch, P. Diaz and N. Nokwara, “Restricted Schur Polynomials for Fermions and integrability in thesu(2|3) sector,” JHEP03(2013) 173 [arXiv:1212.5935 [hep-th]]

    Pith/arXiv arXiv 2013

  29. [29]

    Branes, anti-branes and Brauer algebras in gauge-gravity duality,

    Y. Kimura and S. Ramgoolam, “Branes, anti-branes and Brauer algebras in gauge-gravity duality,” JHEP11(2007) 078 [arXiv:0709.2158 [hep-th]]

    Pith/arXiv arXiv 2007

  30. [30]

    Diagonal multi-matrix correlators and BPS operators inN= 4 SYM,

    T. W. Brown, P. J. Heslop and S. Ramgoolam, “Diagonal multi-matrix correlators and BPS operators inN= 4 SYM,” JHEP02(2008) 030 [arXiv:0711.0176 [hep-th]]

    Pith/arXiv arXiv 2008

  31. [31]

    Diagonal free field matrix correlators, global symmetries and giant gravitons,

    T. W. Brown, P. J. Heslop and S. Ramgoolam, “Diagonal free field matrix correlators, global symmetries and giant gravitons,” JHEP04(2009) 089 [arXiv:0806.1911 [hep-th]]

    Pith/arXiv arXiv 2009

  32. [32]

    The Hagedorn transition, deconfinement andN= 4 SYM theory,

    B. Sundborg, “The Hagedorn transition, deconfinement andN= 4 SYM theory,” Nucl. Phys. B573(2000) 349–363 [hep-th/9908001]

    Pith/arXiv arXiv 2000

  33. [33]

    A pendant for Polya: The one-loop partition function of N= 4 SYM onR×S 3,

    M. Spradlin and A. Volovich,“A pendant for Polya: The one-loop partition function of N= 4 SYM onR×S 3,” Nucl. Phys. B711(2005) 199–230 [hep-th/0408178]. – 39 –

    Pith/arXiv arXiv 2005

  34. [34]

    On stringyAdS 5 ×S 5 and higher spin holography,

    M. Bianchi, J. F. Morales and H. Samtleben, “On stringyAdS 5 ×S 5 and higher spin holography,” JHEP07(2003) 062 [hep-th/0305052]

    Pith/arXiv arXiv 2003

  35. [35]

    Invasion of the giant gravitons from Anti-de Sitter space,

    J. McGreevy, L. Susskind and N. Toumbas, “Invasion of the giant gravitons from Anti-de Sitter space,” JHEP06(2000), 008 [arXiv:hep-th/0003075 [hep-th]]

    Pith/arXiv arXiv 2000

  36. [36]

    Thresholds of large N factorization in CFT 4: exploring bulk spacetime in AdS 5,

    D. Garner, S. Ramgoolam and C. Wen, “Thresholds of large N factorization in CFT 4: exploring bulk spacetime in AdS 5,” JHEP11(2014), 076 [arXiv:1403.5281 [hep-th]]

    Pith/arXiv arXiv 2014

  37. [37]

    The finite group velocity of quantum spin systems,

    E. H. Lieb and D. W. Robinson, “The finite group velocity of quantum spin systems,” Commun. Math. Phys.28(1972), 251-257

    1972

  38. [38]

    Spectral gap and exponential decay of correlations,

    M. B. Hastings and T. Koma, “Spectral gap and exponential decay of correlations,” Commun. Math. Phys.265(2006), 781-804 [arXiv:math-ph/0507008 [math-ph]]

    Pith/arXiv arXiv 2006

  39. [39]

    Locality in Quantum Systems,

    M. B. Hastings, “Locality in Quantum Systems,” [arXiv:1008.5137 [math-ph]]

    Pith/arXiv arXiv

  40. [40]

    Localized shocks,

    D. A. Roberts, D. Stanford and L. Susskind, “Localized shocks,” JHEP03(2015), 051 [arXiv:1409.8180 [hep-th]]. – 40 –

    Pith/arXiv arXiv 2015