Challenges
Datasets
Workspaces
Papers
Discussions
Leaderboard
Log inSign up
Challenges
Datasets
Workspaces
Papers
Discussions
Leaderboard
Blog
Job Board
Q3AS

© 2026 Aqora Quantum S.A.S.

TermsPrivacyLegal Notice

Crossed Cosmos / ECI v7.5.0 — β-scan SU(2) 4D lattice EE sub-leading C-function

Kévin Rémondière

Published May 25, 2026• 0 citations
Publisher

Abstract

⚠️ CORRECTION NOTICE (2026-07-12) - All Table 1 values (c₃D(L,β) ∈ [0.109,0.140] across 132 runs) and the r_c/r_{1/a²}=0.218 ratio-test were produced by driver code that executes, unmodified, the same Metropolis kernel confirmed elsewhere to contain two independent bugs (SU(2) trace sign error K† vs K; reversed staple order), which prevent thermalization to the intended Wilson ensemble. No ⟨plaquette⟩ thermalization table is published anywhere in this record to rule this out.- This record was logged internally as "INVALIDE" the day of its own publication; no public correction followed in the 7 weeks since.- This record's own headline conclusion — that the sub-leading C-function is non-universal in β, correcting v7.4.0's overreach — cannot currently be distinguished from an artifact of the shared broken sampler. The comparison to v7.4.0 (DOI 10.5281/zenodo.20377620) does not resolve this: that record is affected by the identical bug (see its own v2 notice).- No results in this record should be cited as physical measurements of SU(2) lattice entanglement entropy pending a re-run with the corrected kernel.``` Cross-(L, beta) measurement of the sub-leading entanglement entropy C-function in SU(2) 4D lattice gauge theory via Buividovich-Polikarpov alpha-integration. Dataset: beta in {2.3, 2.4, 2.5, 2.6} x L in {8, 10, 12} = 132 simulations. Key findings: 1. c_3D = S_2/(2L^3) grows linearly with beta (slope ~0.10) 2. Ratio test r_c/r_{1/a^2} <= 0.6 proves BP method cancels leading kappa/a^2 3. Sub-leading C(beta) is non-universal (confining SU(2) != CFT) 4. log(3)/(2pi*sqrt(2)) coincidence at beta~2.45 is one-point crossing, not a measurement This paper supersedes the v7.4.0 PRL claim of 'universal C-function = log(3)/(2pi*sqrt(2))' which was based on a single beta=2.4 value. Method limits: does NOT extract leading kappa (cancelled by BP). Roadmap: Modular Hamiltonian, interface free energy, doubled Hilbert space. Author: Kevin Remondiere (ORCID 0009-0008-2443-7166), Independent Researcher, Oloron-Sainte-Marie, France.

DOI
10.5281/zenodo.20378900

Related papers

Crossed Cosmos / ECI v7.3.0 — SU(2) lattice Renyi-2 entanglement entropy BP2008b method validation

Kevin Remondière

⚠️ CORRECTION NOTICE (2026-07-12) - The central result of the abstract and Table I — c₃D(L=12)=0.1215±0.0034, agreeing with log(3)/(2π√2)≈0.1236 at 0.7σ — was produced by the Metropolis kernel deposited with this record (jax_su2_EE_BP2008b_FAST_2026-05-25.py), which contains two independent bugs confirmed the same day this paper was deposited: an SU(2) trace using K† instead of K, and a reversed staple order. A cold-start check on this code gives ⟨plaquette⟩≈−0.18 versus the expected ≈+0.62 at β=2.3 — the sampled configurations never thermalized to the intended Wilson ensemble.- This record was logged internally as "INVALIDE" on its own publication day; that status was never reflected here.- With both bugs fixed, the result is QUALITATIVELY different, not merely refined: κ_EE(SU2)=0.5065±0.010, growing LINEARLY in L (no plateau), corresponding under this paper's own normalization to c₃D≈0.253 — close to the Bekenstein-Hawking value 1/4, not to log(3)/(2π√2). The headline agreement claim is REVERSED.- The companion file PAPER_BP2008b_SU2_EE_JHEP.pdf bundled in this same record inherits the same issue.- The per-link Metropolis speed claim (item 4 of the abstract) is unaffected.``` Cross-lattice measurement of the entanglement entropy C-function in SU(2) 4D lattice gauge theory via Buividovich-Polikarpov alpha-integration. Key result: c_3D = S_2/(2 L_y L_z L_tau) converges monotonically to 0.1222 +/- 0.003 on L = 4,6,8,10,12 at beta=2.4, in 0.7-sigma agreement with the LQG/edge-mode formula log(3)/(2pi*sqrt(2)) = 0.1237 (Donnelly-Wall lattice prediction). Includes: - 4-page PRL letter (PAPER_CFUNCTION_PRL.pdf) - 18-page JHEP method paper (PAPER_BP2008b_SU2_EE_JHEP.pdf) - FAST V3 JAX/GPU code (10x speedup via per-link MH with local staples) - 4 failed methods documented honestly - Analysis script + raw data Method validation pedigree: 4 alternative methods (Donnelly-Wall swap=Jacobien-trivial, plaquette locking, soft A-tying divergent, BAR phase space mismatch) failed before the 5th (BP2008b single-lattice deformed connectivity + Fodor-Endrodi alpha-integration) succeeded. Author: Kevin Remondiere (ORCID 0009-0008-2443-7166), Independent Researcher, Oloron-Sainte-Marie, France.

•May 25, 2026•0 citations

Crossed Cosmos / ECI v7.5.0 — β-scan SU(2) 4D lattice EE sub-leading C-function

Kévin Rémondière

⚠️ CORRECTION NOTICE (2026-11-08; amended 2026-11-09) The original description of this record (27 May 2026) is kept unchanged below. It overstates what the record contains. Read it together with the correction notice of version v10.0.0 of the same concept (10.5281/zenodo.20415354, notice of 2026-07-12) and with the corrections of the repository crossed-cosmos (README, sections "STATUS UPDATE (2026-07-12)" and "INTEGRITY CORRECTION (2026-10-04)"; CHANGELOG, entry 2026-10-04). "World-first G₂ and Sp(4) entanglement entropy" is withdrawn. The priority claims, here and as "first measurement" in the included drafts, are unverified. An internal audit of 2026-06-20 also found that the G₂ result cannot be published as it stands. The G₂/SU(3) ratio given in the drafts (1.291–1.292) changes when both groups are matched at another lattice spacing. Its constancy for L = 4–12 is a stability in volume at fixed couplings, not in the lattice spacing. Part of the stored data could not be reproduced. The attribution of differences between groups to the centre (Donnelly–Wall) is an interpretation that was not tested; the code in this record makes no reference to the centre of the group. "Discovery: S₂/A/β is a Dynkin invariant" is withdrawn. It names an ordering of six numbers, one per group (file pysr_6groups_data.json, key "slopes"). The record does not show that any property of the Dynkin diagram determines them, and the two drafts included in this record do not contain the claim. The values in that file order the groups A₁ > B₃ > C₂ > A₂ > A₃ > G₂. The description of v10.0.0 states C₂ > B₃ > A₁ > A₂ > A₃ > G₂. The PySR formula is a fit, not a discovery or a derivation. Its coefficients were fitted a posteriori to the measured values of S₂/A by symbolic regression. The internal record of the fit also has an additive constant (+2.97) that the description omits. R² = 0.999 measures the quality of that fit; it is not a test on independent data. The v10 notice requalifies the later formula S₂/A = (π+e)(β−|Φ⁺|) − log(β−1−|Φ⁺|) − C₂ in the same way: (π+e) is a fitted constant, not a zero-parameter derivation. "Paper PRL 3pp included" refers to Paper_G2_EE_PRL_6groups.pdf. It is a three-page draft in PRL format, not a published PRL paper, and it still has unresolved citations and cross-references. It covers SU(2), SU(3), SU(4) and G₂ only. It contains none of the Sp(4), SO(7), Dynkin or PySR statements of the description. "Sp(4) falsifies simple C₂+ln|Z| at 50%" is unverified. The internal notes of 27 May 2026 contradict each other on it (one of them attributes most of the deviation to the choice of couplings). Amended 2026-11-09: the SU(2) β-scan files of this record must not be used. PAPER_BETASCAN_SU2_EE (.tex, .pdf, .pdf.ots), jax_su2_EE_BETASCAN_results.json, jax_su2_EE_BETASCAN_2026-05-25.py, analysis_betascan_2026-05-25.py and jax_su2_EE_BP2008b_FAST_2026-05-25.py are the files of version v7.5.0 (10.5281/zenodo.20378901, also deposited as 10.5281/zenodo.20379361), with the same MD5 checksums. All measured values in them were produced by the Metropolis kernel jax_su2_EE_BP2008b_FAST_2026-05-25.py, which has two independent bugs: its acceptance step uses Re Tr(U·K†) where Re Tr(U·K) is required, and the forward staple has its links in the wrong order. A cold-start test of the kernel (β = 2.4, L = 8) gives ⟨P⟩ ≈ −0.18 instead of about +0.6, so the configurations are not samples of the SU(2) Wilson action. The c₃D(L, β) values, their β-dependence and the claimed cancellation of the leading area-law term are therefore not physical results. The bugs were fixed in the repository crossed-cosmos on 25 May 2026 (commits 40346b0 and 73fbc43), before this record was published, but these files were included without being re-computed, and no corrected β-scan appears in the author's records. See also the notice of 2026-07-12 on 10.5281/zenodo.20378901. This notice is not a full audit of the record. The code, the 46 data points and the SU(2) β-scan draft (PAPER_BETASCAN_SU2_EE) are not re-checked here. Amended 2026-11-09: this notice is not a full audit of the record; the code and the 46 data points are not re-checked here (for the SU(2) β-scan files, see the point above). No file of the record has been changed. Original description (27 May 2026), unchanged: Kevinotron lattice gauge theory EE engine: 46 data points across SU(2), SU(3), SU(4), G₂, Sp(4), SO(7). World-first G₂ and Sp(4) entanglement entropy. Discovery: S₂/A/β is a Dynkin invariant. PySR formula: S₂/A ≈ 5.85·(β−|Φ⁺|) − 2.23·C₂ (R²=0.999). Sp(4) falsifies simple C₂+ln|Z| at 50%. Paper PRL 3pp included.

•May 25, 2026•0 citations

NHJQ-SK final project record (v3) The J-Q model under an imaginary Dzyaloshinskii–Moriya perturbation: mechanism, sign-free validation, the structure of the discrimination problem

Kamil Vargovský, Maria Vargovská

Non-Hermitian J-Q model in the SU(2) limit: S^x channel reframing of Q_c(δ) (v2) Authors: Kamil Vargovský, Mária Vargovská Date: 15 May 2026 Supersedes: v1 (DOI 10.5281/zenodo.19451609, 7 April 2026) HPC allocation: Devana cluster, NSCC SAV, project p1927-26-t Changelog v1 → v2 Version 1 (April 2026) reported a threshold-monotonic Q_c(δ) interpretation based on S^z structure factor measurements. Subsequent analysis identified S^z as an inappropriate observable in the non-Hermitian (NH) regime due to broken rotational symmetry under δ ≠ 0. The threshold interpretation in v1 is superseded and obsolete, not refined or extended, by this v2. Quantitative v1 results remain valid as S^z-channel-specific observations within the measured range but should not be cited as universal phase-boundary characterization for the non-Hermitian J-Q model. Context The deconfined quantum critical point (DQCP) separating Néel antiferromagnetic and valence bond solid phases challenges the Landau-Ginzburg-Wilson framework. Zou et al. (arXiv:2511.03456, 2025) showed that NH perturbations weaken first-order tendencies at the DQCP in the easy-plane J-Q model (Δ=0.6). This work extends to the SU(2) limit (Δ=0) using sign-free SSE QMC on the square lattice with periodic boundary conditions. The S^x transverse channel is probed via an off-diagonal correlation estimator (Weber PhD §2.8 framework based), L ∈ {8, 12, 16, 24, 32, 48, 64}, δ ∈ [0, 1], Q ∈ [6, 40], T = 0.1. Full methodology reserved for follow-up publication. Key Findings Finding 1 — H_PHASE_STABLE falsified with qualitative L-scaling asymmetry between hermitian and non-Hermitian limits.** At Q=22, S^xx_uniform exhibits linear-in-(1/L) decay at δ=0 across L ∈ {16, 24, 32, 48, 64} (values 0.01112, 0.01013, 0.00964, 0.00859, 0.00806; L=48 anchor matches linear interpolation between L=32 and L=64 within 0.02σ), consistent with disordered-phase asymptotic scaling. At δ=1, the same L range shows monotonic growth with super-linear deviation from linear FSS (values 0.01170, 0.01131, 0.01125, 0.01307, 0.01556; the L=64 entry is the mean of 3 independent replica measurements at seeds {default, 2026051501, 2026051502}, SEM = 1.90×10⁻⁴, inter-replica scatter <2.1σ between any pair). The L=48 anchor at δ=1 lies 3.86σ below the linear-in-(1/L) interpolation between L=32 and L=64, confirming non-linear FSS deviation independent of L=64 single-measurement concerns. The L=64 ratio S^xx_uniform(δ=1)/S^xx_uniform(δ=0) = 1.931 ± 0.027 demonstrates 3-point L-scaling asymmetry (decay δ=0 vs accelerating growth δ=1) statistically robust at 20.2σ for δ=1 (L=32→L=64 transition) and 6.27σ for δ=0 (L=32→L=64 transition), independently established across 5 lattice sizes. Finding 2 — Sign-free preservation across 56 production configurations** spanning δ ∈ [0,1], Q ∈ [6,40], L ∈ [4,64]. Exact unity for ⟨sign⟩ and SignFracPos diagnostics confirms the analytical Marshall boundary |δ| ≤ 1 for the local per-bond decomposition. Finding 3 — No S^xx_uniform peak detected at δ ∈ {0.6, 1.0} within the measured Q range.** Monotonic decreasing behavior across Q ∈ [14, 40] at δ=0.6 and Q ∈ [6, 28] at δ=1.0. Different from the naïve Sandvik phase diagram [Sandvik 2007, arXiv:cond-mat/0611343] in the S^x channel within the measured Q range. Finding 4 — Interior minimum hypothesis for Q*(δ) crossover trajectory empirically falsified.** The L=24/L=16 ratio across an extended 6-point δ-scan yields a non-decreasing trajectory with a plateau between δ=0.3 and δ=0.5 within quoted precision. The interior minimum hypothesis, motivated by an initial 3-point δ-anchor (δ ∈ {0, 0.6, 1.0}), is empirically falsified by the 6-point measurement (δ ∈ {0, 0.3, 0.5, 0.6, 0.7, 1.0}). Interpretation (requires Phase 3) Whether the super-linear FSS deviation at δ=1 signals a phase transition crossover, the emergence of a distinct phase in the NH thermodynamic limit, or a Marshall local convention edge case, remains under investigation. The 3-replica statistical robustness at L=64 and L=48 anchor consistency reduce single-measurement concerns but do not resolve the underlying mechanism. Phase 3 outlook L=96 measurement for η extraction; dimer-dimer alternative order parameter cross-validation; sweep count scaling for calibration gap discrimination. Keywords: deconfined quantum criticality, non-Hermitian quantum mechanics, J-Q model, SSE QMC, finite-size scaling, sign-free Monte Carlo, SU(2) symmetry Subject areas: Condensed Matter Physics, Quantum Many-Body Systems, Computational Physics References 1. Zou, X., Yin, S., Li, Z.-X., Yao, H. "Unraveling Deconfined Quantum Criticality in Non-Hermitian Easy-Plane J-Q Model." arXiv:2511.03456 (2025). 2. Sandvik, A. W. "Evidence for deconfined quantum criticality in a two-dimensional Heisenberg model with four-spin interactions." Phys. Rev. Lett. 98, 227202 (2007); arXiv:cond-mat/0611343. 3. Senthil, T. et al. "Deconfined quantum critical points." Science 303, 1490 (2004). 4. Weber, L. "Quantum Monte Carlo methods for critical quantum magnets." PhD thesis, RWTH Aachen University, 2022. This v2 supersedes the v1 (April 2026) interpretation. Full methodology, raw dataset, and production code are reserved for follow-up preprint publication. Technical correspondence on methodology and findings is welcome.

•Apr 7, 2026•0 citations

Anyon Condensation in Modular Tensor Categories: Computational Classification and Algebraic Structure (v4)

Adam Bazih

Author note (Nov 2026): an independent re-derivation confirms the two-algebra classification, the SO(11)_1 condensed phase, the N+k=6 characterization and the Witt class. It also found errors to be fixed in v8: the v7 erratum wrongly removed the SU(3)_2 x SU(2)_3 algebra, which is the SU(6)_1 conformal embedding (Theorem 4.1 should read 6 of 10, one E-type); mtctools 0.3.0 uses Re(S) in its invariant routine and accepts unphysical invariants (SO(9)_2 "d_A = 4"); WITT_TABLE central charges for F4_2 (16/11) and E6_2 (22/7) were rounded; the zero-locus conjecture fails for SU(2)_7. Version 7 — strengthens and corrects v6. 1. Adjoint transparency, characterized and explained. The v6 statement ("unique to SU(3)_3 among SU(N)_N") is superseded twice over. (a) Level-rank characterization: for SU(N)_k, k>=2, S_adj,fund = 0 if and only if N+k=6 (SU(2)_4, SU(3)_3, SU(4)_2 — one level-rank orbit), proved in closed form. (b) Mechanism and cross-family classification: the adjoint is a simple-current FIXED POINT exactly for SU(N)_{6-N} and for ALL Spin(N)_2, N>=5; a two-line lemma (multiplicativity of Verlinde eigenvalues) then forces S_adj,x = 0 for every current-charged x. The previously "isolated" Sp(2)_2 = Spin(5)_2 is the first member of the infinite metaplectic family. All S-matrix zeros in the scanned families are accounted for by this mechanism or by known exceptional invariants (SU(3)_5 in SU(6)_1, G2 level 3 in E6 level 1, ...); we conjecture no accidental zeros exist in WZW modular data. 2. Corrected survey. (a) The earlier non-simple-current algebra in SU(3)_2 ⊗ SU(2)_3 does not exist (non-unitary S-matrix artifact). (b) Condensed phases are now identified by rank+twists+central charge, never by D^2 alone: the Z2 condensation of SU(2)_2 ⊗ SU(2)_2 is the pointed theory SU(4)_1 (nu=6), NOT Ising; the maximal condensation of SU(3)_3 ⊗ SU(2)_2 is the nu=11 Ising-type category of Kitaev's sixteen-fold way — the modular tensor category of SO(11)_1 (h_sigma = 11/16), which shares its fusion rules with Ising and its fusion category with SU(2)_2 but is equivalent to neither as a braided category. Its Witt class is [Ising]^11, of order 16. 3. mtctools 0.3.0. The condensable-algebra finder now uses the modular-invariant (commutant) criterion decided by exact integer programming; the previous z_l-integrality test had false negatives on conformal-embedding algebras (it rejected 1+(j=3) in SU(2)_10, 1+(2,2) in SU(3)_5, 1+(1,1) in G2_3 — all now found and covered by regression tests). Phase identification is sixteen-fold-aware (takes c mod 8). Two further fixes: a twist-rounding bug that discarded B-series algebras (including the LAGRANGIAN of SO(9)_2 — which has c=0 mod 8 and a gapped boundary), and fixed-point splitting is now restricted to cyclic stabilizers (non-cyclic cases are flagged; they require discrete-torsion data beyond (S,T)). 4. New data: a Witt-class table for the metaplectic series SO(N)_2 (N=5..13), whose vector-current condensates are pointed (rank = D^2 forces it), reducing each Witt class to an explicit finite quadratic form (e.g. (Z_11, k^2/11), (Z_13, 2k^2/13)); SO(9)_2 is Witt-trivial. See ERRATUM_v7.md for the changelog. All computations reproducible with the included package (S unitary, S^2=C, integer Verlinde, ~1e-15).

•May 12, 2026•0 citations

ECI v6 — From Faulkner-Speranza to a complexity-bounded GSL on type-II crossed products

Kevin Remondière

⚠️ CORRECTION NOTICE (2026-07-12) - The boxed "Main result (Theorem)" and the abstract's "We prove an upper bound" overstate what is established: the bound depends on Assumption M1, explicitly labeled a POSTULATE (modular complexity ansatz) in this paper's own text. A contemporaneous internal multi-agent audit (3/3 independent reviewers) concluded the M1→Jacobson-EGJ bridge has "no route currently known" and recommended explicitly: "do not claim derivation." Recommend downgrading "Theorem"/"prove" to "conjectural bound under Assumption M1" in the abstract and result box.- Two papers extending Krylov complexity into the modular/horizon direction flagged as "surveillance targets" in the pre-submission novelty audit have since appeared (arXiv:2507.23739, arXiv:2602.02675); neither reproduces this paper's full composite equation, so no scoop has occurred, but the novelty margin is narrowing while this remains unsubmitted to arXiv.- This record is dormant (no update since May 2026) but not shown to be false; it sits outside the scope of the current lattice/mass-gap programme and should be read as such.``` We prove an upper bound on the rate of generalised entropy along the modular (Connes-Tomita-Takesaki) flow τ_R of a type-II crossed-product observer algebra A_R, extending the Eling-Guedens-Jacobson local Clausius schema dS = δQ/T + d_iS to the setting of Chandrasekaran-Longo-Penington-Witten and De Vuyst-Eccles-Held-Kirklin. The bound reads dS_gen[R]/dτ_R ≤ κ_R · C_k[ρ_R(τ_R)] · Θ(PH_k[δn(τ_R)]) with κ_R = 2π T_R the Bisognano-Wichmann modular surface gravity, C_k a k-design / spread complexity proxy, and Θ a topological activator on persistent homology of the dequantised classical density field δn. v6.0.16 retracts the v6.0.15 framing of M1 as a "conditional theorem" — see the framework paper (Remondière, ECI v6.0.16, DOI 10.5281/zenodo.19708889) and notes/m1_audit_2026_05_02.md. M1 retains POSTULATE status; the v8-bis math.CT companion target is Conjecture M1-C.

•May 2, 2026•0 citations

[SUPERSEDED DUPLICATE of 10.5281/zenodo.21698294] Magic-State Factory Performance for Single-Shot 4D Toric Codes: the Factory Floor Dominates

James Jardine

SUPERSEDED DUPLICATE — do not cite this record. Cite 10.5281/zenodo.21698294 instead. Both records were deposited on 2026-07-30 and contain the same paper. This one carries an earlier, superseded NOTE.md whose resource-scaling section still claimed L≈6–7 (d=36–49); the surviving record carries the corrected text (measured L=4, [[1536,6,16]]) and the toric4d_lowp.py stratified estimator behind it. The independent cross-validation scripts that were unique to this record (magic_state4-7.py) have been added to the surviving record. Flagged 2026-07-31. We simulate a 15-to-1 magic-state factory under full circuit-level depolarizing noise and ask whether a 4D toric code — which corrects errors single-shot, without repeated syndrome rounds — relieves the factory's dominant cost. It does not. Distillation output error is confirmed at 35p³ by exact enumeration and by stim, but the factory's circuit-level error floor is first order in the gate error rate and schedule-dependent, measured at 12–15·p_gate for the real [[15,1,3]] quantum Reed–Muller code with all 14 stabilizers. This is 3.5× larger than a previously reported 4.18·p_gate, which we trace to a circuit that omitted the code's 10 Z-type checks; that figure is withdrawn here. The floor originates in the final syndrome sweep: faults deposited on data qubits by the last round of gates are never checked by anything after them. Whichever check type is measured last, that sector's faults escape, so rescheduling shifts the floor between sectors and interleaving shaves roughly 20%, but cannot remove it. Reordering, redundant checks and flag qubits leave it first order; decoding weight-1 errors instead of discarding them makes it 5× worse. We construct the 4D toric code on an L⁴ torus from its cubical chain complex and verify [[486,6,9]] at L=3 (k=6 exactly, explicit weight-9 logicals in both sectors, d=4 exact at L=2 by exhaustive search), and measure a code-capacity threshold between 5% and 7% under BP+OSD. Single-shot repair using the 80 independent metachecks recovers near-perfect-syndrome performance, but only while syndrome noise stays roughly 10–30× below the data error rate: it improves logical error by 195× at q=10⁻³ and by only 4.5× at q=10⁻². Crucially, metachecks constrain the syndrome, not the data, and cannot see faults deposited directly on data qubits by the final sweep — which is precisely where the factory floor originates. Single-shot therefore buys latency, not factory fidelity. Because the factory's Cliffords act on encoded blocks, p_gate is the logical error rate of the hosting code. The correction moves the surface-code requirement from d=15 to d=17 (578 physical qubits per patch); on the 4D toric side, the smallest sufficient block is L=4, [[1536,6,16]], with [[486,6,9]] missing the fidelity demand by four to five orders of magnitude. We situate the result against Xu et al. (arXiv:2407.18490), whose parallel distillation across qLDPC patches reduces space–time cost but leaves the per-factory floor untouched. A corrections log records eight claims withdrawn or corrected during this work, including our own earlier 4.18·p_gate figure and a d=15 requirement that rested on a single error event. Section 6 states what is not established: these are code-capacity numbers for the 4D code, the distance requirement is quoted at p=3–5% rather than 10⁻³, d=L² is verified exactly only at L=2, the Λ extrapolations are long, and no end-to-end injection simulation was run. Correction, 2026-07-31. The files deposited with this record on 2026-07-30 predate a low-p re-measurement of the resource-scaling section. The originally deposited text stated that hosting the factory requires L≈6–7 (d=36–49, n≈7800–14400); that figure came from fitting a single suppression factor Λ to near-threshold points at p=3–5%. The single-Λ ansatz does not hold for this code: the measured suppression per two units of distance runs from 10.9 (d=4→9 at p=0.02) to 793 (d=9→16 at p=0.01). Re-measuring at the factory's actual operating regime with a weight-stratified estimator gives d ≥ 14.5 at p=0.02 and d ≥ 9.7 at p=0.01, so the smallest sufficient 4D block is L=4, [[1536,6,16]], not L=6–7. The L≈6–7 figure is withdrawn. These are point estimates whose upper confidence bands remain wide at the smallest p, and should be read as order-of-magnitude. The corrected text and the estimator script (toric4d_lowp.py) are not in this record's file set; replacing files requires a new version, which will carry the corrected note. Nothing in Sections 1–4 — the 35p³ distillation result, the 12–15·p_gate factory floor, the last-sweep mechanism, or the d=17 surface-code requirement — is affected.

•Jul 30, 2026•0 citations
⚠️ CORRECTION NOTICE (2026-07-12) - All Table 1 values (c₃D(L,β) ∈ [0.109,0.140] across 132 runs) and the r_c/r_{1/a²}=0.218 ratio-test were produced by driver code that executes, unmodified, the same Metropolis kernel confirmed elsewhere to contain two independent bugs (SU(2) trace sign error K† vs K; reversed staple order), which prevent thermalization to the intended Wilson ensemble. No ⟨plaquette⟩ thermalization table is published anywhere in this record to rule this out.- This record was logged internally as "INVALIDE" the day of its own publication; no public correction followed in the 7 weeks since.- This record's own headline conclusion — that the sub-leading C-function is non-universal in β, correcting v7.4.0's overreach — cannot currently be distinguished from an artifact of the shared broken sampler. The comparison to v7.4.0 (DOI 10.5281/zenodo.20377620) does not resolve this: that record is affected by the identical bug (see its own v2 notice).- No results in this record should be cited as physical measurements of SU(2) lattice entanglement entropy pending a re-run with the corrected kernel.``` Cross-(L, beta) measurement of the sub-leading entanglement entropy C-function in SU(2) 4D lattice gauge theory via Buividovich-Polikarpov alpha-integration. Dataset: beta in {2.3, 2.4, 2.5, 2.6} x L in {8, 10, 12} = 132 simulations. Key findings: 1. c_3D = S_2/(2L^3) grows linearly with beta (slope ~0.10) 2. Ratio test r_c/r_{1/a^2} <= 0.6 proves BP method cancels leading kappa/a^2 3. Sub-leading C(beta) is non-universal (confining SU(2) != CFT) 4. log(3)/(2pisqrt(2)) coincidence at beta~2.45 is one-point crossing, not a measurement This paper supersedes the v7.4.0 PRL claim of 'universal C-function = log(3)/(2pisqrt(2))' which was based on a single beta=2.4 value. Method limits: does NOT extract leading kappa (cancelled by BP). Roadmap: Modular Hamiltonian, interface free energy, doubled Hilbert space. Author: Kevin Remondiere (ORCID 0009-0008-2443-7166), Independent Researcher, Oloron-Sainte-Marie, France.
⚠️ CORRECTION NOTICE (2026-07-12) - The central result of the abstract and Table I — c₃D(L=12)=0.1215±0.0034, agreeing with log(3)/(2π√2)≈0.1236 at 0.7σ — was produced by the Metropolis kernel deposited with this record (jax_su2_EE_BP2008b_FAST_2026-05-25.py), which contains two independent bugs confirmed the same day this paper was deposited: an SU(2) trace using K† instead of K, and a reversed staple order. A cold-start check on this code gives ⟨plaquette⟩≈−0.18 versus the expected ≈+0.62 at β=2.3 — the sampled configurations never thermalized to the intended Wilson ensemble.- This record was logged internally as "INVALIDE" on its own publication day; that status was never reflected here.- With both bugs fixed, the result is QUALITATIVELY different, not merely refined: κ_EE(SU2)=0.5065±0.010, growing LINEARLY in L (no plateau), corresponding under this paper's own normalization to c₃D≈0.253 — close to the Bekenstein-Hawking value 1/4, not to log(3)/(2π√2). The headline agreement claim is REVERSED.- The companion file PAPER_BP2008b_SU2_EE_JHEP.pdf bundled in this same record inherits the same issue.- The per-link Metropolis speed claim (item 4 of the abstract) is unaffected.``` Cross-lattice measurement of the entanglement entropy C-function in SU(2) 4D lattice gauge theory via Buividovich-Polikarpov alpha-integration. Key result: c_3D = S_2/(2 L_y L_z L_tau) converges monotonically to 0.1222 +/- 0.003 on L = 4,6,8,10,12 at beta=2.4, in 0.7-sigma agreement with the LQG/edge-mode formula log(3)/(2pi*sqrt(2)) = 0.1237 (Donnelly-Wall lattice prediction). Includes: - 4-page PRL letter (PAPER_CFUNCTION_PRL.pdf) - 18-page JHEP method paper (PAPER_BP2008b_SU2_EE_JHEP.pdf) - FAST V3 JAX/GPU code (10x speedup via per-link MH with local staples) - 4 failed methods documented honestly - Analysis script + raw data Method validation pedigree: 4 alternative methods (Donnelly-Wall swap=Jacobien-trivial, plaquette locking, soft A-tying divergent, BAR phase space mismatch) failed before the 5th (BP2008b single-lattice deformed connectivity + Fodor-Endrodi alpha-integration) succeeded. Author: Kevin Remondiere (ORCID 0009-0008-2443-7166), Independent Researcher, Oloron-Sainte-Marie, France.
⚠️ CORRECTION NOTICE (2026-11-08; amended 2026-11-09) The original description of this record (27 May 2026) is kept unchanged below. It overstates what the record contains. Read it together with the correction notice of version v10.0.0 of the same concept (10.5281/zenodo.20415354, notice of 2026-07-12) and with the corrections of the repository crossed-cosmos (README, sections "STATUS UPDATE (2026-07-12)" and "INTEGRITY CORRECTION (2026-10-04)"; CHANGELOG, entry 2026-10-04). "World-first G₂ and Sp(4) entanglement entropy" is withdrawn. The priority claims, here and as "first measurement" in the included drafts, are unverified. An internal audit of 2026-06-20 also found that the G₂ result cannot be published as it stands. The G₂/SU(3) ratio given in the drafts (1.291–1.292) changes when both groups are matched at another lattice spacing. Its constancy for L = 4–12 is a stability in volume at fixed couplings, not in the lattice spacing. Part of the stored data could not be reproduced. The attribution of differences between groups to the centre (Donnelly–Wall) is an interpretation that was not tested; the code in this record makes no reference to the centre of the group. "Discovery: S₂/A/β is a Dynkin invariant" is withdrawn. It names an ordering of six numbers, one per group (file pysr_6groups_data.json, key "slopes"). The record does not show that any property of the Dynkin diagram determines them, and the two drafts included in this record do not contain the claim. The values in that file order the groups A₁ > B₃ > C₂ > A₂ > A₃ > G₂. The description of v10.0.0 states C₂ > B₃ > A₁ > A₂ > A₃ > G₂. The PySR formula is a fit, not a discovery or a derivation. Its coefficients were fitted a posteriori to the measured values of S₂/A by symbolic regression. The internal record of the fit also has an additive constant (+2.97) that the description omits. R² = 0.999 measures the quality of that fit; it is not a test on independent data. The v10 notice requalifies the later formula S₂/A = (π+e)(β−|Φ⁺|) − log(β−1−|Φ⁺|) − C₂ in the same way: (π+e) is a fitted constant, not a zero-parameter derivation. "Paper PRL 3pp included" refers to Paper_G2_EE_PRL_6groups.pdf. It is a three-page draft in PRL format, not a published PRL paper, and it still has unresolved citations and cross-references. It covers SU(2), SU(3), SU(4) and G₂ only. It contains none of the Sp(4), SO(7), Dynkin or PySR statements of the description. "Sp(4) falsifies simple C₂+ln|Z| at 50%" is unverified. The internal notes of 27 May 2026 contradict each other on it (one of them attributes most of the deviation to the choice of couplings). Amended 2026-11-09: the SU(2) β-scan files of this record must not be used. PAPER_BETASCAN_SU2_EE (.tex, .pdf, .pdf.ots), jax_su2_EE_BETASCAN_results.json, jax_su2_EE_BETASCAN_2026-05-25.py, analysis_betascan_2026-05-25.py and jax_su2_EE_BP2008b_FAST_2026-05-25.py are the files of version v7.5.0 (10.5281/zenodo.20378901, also deposited as 10.5281/zenodo.20379361), with the same MD5 checksums. All measured values in them were produced by the Metropolis kernel jax_su2_EE_BP2008b_FAST_2026-05-25.py, which has two independent bugs: its acceptance step uses Re Tr(U·K†) where Re Tr(U·K) is required, and the forward staple has its links in the wrong order. A cold-start test of the kernel (β = 2.4, L = 8) gives ⟨P⟩ ≈ −0.18 instead of about +0.6, so the configurations are not samples of the SU(2) Wilson action. The c₃D(L, β) values, their β-dependence and the claimed cancellation of the leading area-law term are therefore not physical results. The bugs were fixed in the repository crossed-cosmos on 25 May 2026 (commits 40346b0 and 73fbc43), before this record was published, but these files were included without being re-computed, and no corrected β-scan appears in the author's records. See also the notice of 2026-07-12 on 10.5281/zenodo.20378901. This notice is not a full audit of the record. The code, the 46 data points and the SU(2) β-scan draft (PAPER_BETASCAN_SU2_EE) are not re-checked here. Amended 2026-11-09: this notice is not a full audit of the record; the code and the 46 data points are not re-checked here (for the SU(2) β-scan files, see the point above). No file of the record has been changed. Original description (27 May 2026), unchanged: Kevinotron lattice gauge theory EE engine: 46 data points across SU(2), SU(3), SU(4), G₂, Sp(4), SO(7). World-first G₂ and Sp(4) entanglement entropy. Discovery: S₂/A/β is a Dynkin invariant. PySR formula: S₂/A ≈ 5.85·(β−|Φ⁺|) − 2.23·C₂ (R²=0.999). Sp(4) falsifies simple C₂+ln|Z| at 50%. Paper PRL 3pp included.
Non-Hermitian J-Q model in the SU(2) limit: S^x channel reframing of Q_c(δ) (v2) Authors: Kamil Vargovský, Mária Vargovská Date: 15 May 2026 Supersedes: v1 (DOI 10.5281/zenodo.19451609, 7 April 2026) HPC allocation: Devana cluster, NSCC SAV, project p1927-26-t Changelog v1 → v2 Version 1 (April 2026) reported a threshold-monotonic Q_c(δ) interpretation based on S^z structure factor measurements. Subsequent analysis identified S^z as an inappropriate observable in the non-Hermitian (NH) regime due to broken rotational symmetry under δ ≠ 0. The threshold interpretation in v1 is superseded and obsolete, not refined or extended, by this v2. Quantitative v1 results remain valid as S^z-channel-specific observations within the measured range but should not be cited as universal phase-boundary characterization for the non-Hermitian J-Q model. Context The deconfined quantum critical point (DQCP) separating Néel antiferromagnetic and valence bond solid phases challenges the Landau-Ginzburg-Wilson framework. Zou et al. (arXiv:2511.03456, 2025) showed that NH perturbations weaken first-order tendencies at the DQCP in the easy-plane J-Q model (Δ=0.6). This work extends to the SU(2) limit (Δ=0) using sign-free SSE QMC on the square lattice with periodic boundary conditions. The S^x transverse channel is probed via an off-diagonal correlation estimator (Weber PhD §2.8 framework based), L ∈ {8, 12, 16, 24, 32, 48, 64}, δ ∈ [0, 1], Q ∈ [6, 40], T = 0.1. Full methodology reserved for follow-up publication. Key Findings Finding 1 — H_PHASE_STABLE falsified with qualitative L-scaling asymmetry between hermitian and non-Hermitian limits.** At Q=22, S^xx_uniform exhibits linear-in-(1/L) decay at δ=0 across L ∈ {16, 24, 32, 48, 64} (values 0.01112, 0.01013, 0.00964, 0.00859, 0.00806; L=48 anchor matches linear interpolation between L=32 and L=64 within 0.02σ), consistent with disordered-phase asymptotic scaling. At δ=1, the same L range shows monotonic growth with super-linear deviation from linear FSS (values 0.01170, 0.01131, 0.01125, 0.01307, 0.01556; the L=64 entry is the mean of 3 independent replica measurements at seeds {default, 2026051501, 2026051502}, SEM = 1.90×10⁻⁴, inter-replica scatter <2.1σ between any pair). The L=48 anchor at δ=1 lies 3.86σ below the linear-in-(1/L) interpolation between L=32 and L=64, confirming non-linear FSS deviation independent of L=64 single-measurement concerns. The L=64 ratio S^xx_uniform(δ=1)/S^xx_uniform(δ=0) = 1.931 ± 0.027 demonstrates 3-point L-scaling asymmetry (decay δ=0 vs accelerating growth δ=1) statistically robust at 20.2σ for δ=1 (L=32→L=64 transition) and 6.27σ for δ=0 (L=32→L=64 transition), independently established across 5 lattice sizes. Finding 2 — Sign-free preservation across 56 production configurations** spanning δ ∈ [0,1], Q ∈ [6,40], L ∈ [4,64]. Exact unity for ⟨sign⟩ and SignFracPos diagnostics confirms the analytical Marshall boundary |δ| ≤ 1 for the local per-bond decomposition. Finding 3 — No S^xx_uniform peak detected at δ ∈ {0.6, 1.0} within the measured Q range.** Monotonic decreasing behavior across Q ∈ [14, 40] at δ=0.6 and Q ∈ [6, 28] at δ=1.0. Different from the naïve Sandvik phase diagram [Sandvik 2007, arXiv:cond-mat/0611343] in the S^x channel within the measured Q range. Finding 4 — Interior minimum hypothesis for Q*(δ) crossover trajectory empirically falsified.** The L=24/L=16 ratio across an extended 6-point δ-scan yields a non-decreasing trajectory with a plateau between δ=0.3 and δ=0.5 within quoted precision. The interior minimum hypothesis, motivated by an initial 3-point δ-anchor (δ ∈ {0, 0.6, 1.0}), is empirically falsified by the 6-point measurement (δ ∈ {0, 0.3, 0.5, 0.6, 0.7, 1.0}). Interpretation (requires Phase 3) Whether the super-linear FSS deviation at δ=1 signals a phase transition crossover, the emergence of a distinct phase in the NH thermodynamic limit, or a Marshall local convention edge case, remains under investigation. The 3-replica statistical robustness at L=64 and L=48 anchor consistency reduce single-measurement concerns but do not resolve the underlying mechanism. Phase 3 outlook L=96 measurement for η extraction; dimer-dimer alternative order parameter cross-validation; sweep count scaling for calibration gap discrimination. Keywords: deconfined quantum criticality, non-Hermitian quantum mechanics, J-Q model, SSE QMC, finite-size scaling, sign-free Monte Carlo, SU(2) symmetry Subject areas: Condensed Matter Physics, Quantum Many-Body Systems, Computational Physics References 1. Zou, X., Yin, S., Li, Z.-X., Yao, H. "Unraveling Deconfined Quantum Criticality in Non-Hermitian Easy-Plane J-Q Model." arXiv:2511.03456 (2025). 2. Sandvik, A. W. "Evidence for deconfined quantum criticality in a two-dimensional Heisenberg model with four-spin interactions." Phys. Rev. Lett. 98, 227202 (2007); arXiv:cond-mat/0611343. 3. Senthil, T. et al. "Deconfined quantum critical points." Science 303, 1490 (2004). 4. Weber, L. "Quantum Monte Carlo methods for critical quantum magnets." PhD thesis, RWTH Aachen University, 2022. This v2 supersedes the v1 (April 2026) interpretation. Full methodology, raw dataset, and production code are reserved for follow-up preprint publication. Technical correspondence on methodology and findings is welcome.
Author note (Nov 2026): an independent re-derivation confirms the two-algebra classification, the SO(11)_1 condensed phase, the N+k=6 characterization and the Witt class. It also found errors to be fixed in v8: the v7 erratum wrongly removed the SU(3)_2 x SU(2)_3 algebra, which is the SU(6)_1 conformal embedding (Theorem 4.1 should read 6 of 10, one E-type); mtctools 0.3.0 uses Re(S) in its invariant routine and accepts unphysical invariants (SO(9)_2 "d_A = 4"); WITT_TABLE central charges for F4_2 (16/11) and E6_2 (22/7) were rounded; the zero-locus conjecture fails for SU(2)_7. Version 7 — strengthens and corrects v6. 1. Adjoint transparency, characterized and explained. The v6 statement ("unique to SU(3)_3 among SU(N)_N") is superseded twice over. (a) Level-rank characterization: for SU(N)_k, k>=2, S_adj,fund = 0 if and only if N+k=6 (SU(2)_4, SU(3)_3, SU(4)_2 — one level-rank orbit), proved in closed form. (b) Mechanism and cross-family classification: the adjoint is a simple-current FIXED POINT exactly for SU(N)_{6-N} and for ALL Spin(N)_2, N>=5; a two-line lemma (multiplicativity of Verlinde eigenvalues) then forces S_adj,x = 0 for every current-charged x. The previously "isolated" Sp(2)_2 = Spin(5)_2 is the first member of the infinite metaplectic family. All S-matrix zeros in the scanned families are accounted for by this mechanism or by known exceptional invariants (SU(3)_5 in SU(6)_1, G2 level 3 in E6 level 1, ...); we conjecture no accidental zeros exist in WZW modular data. 2. Corrected survey. (a) The earlier non-simple-current algebra in SU(3)_2 ⊗ SU(2)_3 does not exist (non-unitary S-matrix artifact). (b) Condensed phases are now identified by rank+twists+central charge, never by D^2 alone: the Z2 condensation of SU(2)_2 ⊗ SU(2)_2 is the pointed theory SU(4)_1 (nu=6), NOT Ising; the maximal condensation of SU(3)_3 ⊗ SU(2)_2 is the nu=11 Ising-type category of Kitaev's sixteen-fold way — the modular tensor category of SO(11)_1 (h_sigma = 11/16), which shares its fusion rules with Ising and its fusion category with SU(2)_2 but is equivalent to neither as a braided category. Its Witt class is [Ising]^11, of order 16. 3. mtctools 0.3.0. The condensable-algebra finder now uses the modular-invariant (commutant) criterion decided by exact integer programming; the previous z_l-integrality test had false negatives on conformal-embedding algebras (it rejected 1+(j=3) in SU(2)_10, 1+(2,2) in SU(3)_5, 1+(1,1) in G2_3 — all now found and covered by regression tests). Phase identification is sixteen-fold-aware (takes c mod 8). Two further fixes: a twist-rounding bug that discarded B-series algebras (including the LAGRANGIAN of SO(9)_2 — which has c=0 mod 8 and a gapped boundary), and fixed-point splitting is now restricted to cyclic stabilizers (non-cyclic cases are flagged; they require discrete-torsion data beyond (S,T)). 4. New data: a Witt-class table for the metaplectic series SO(N)_2 (N=5..13), whose vector-current condensates are pointed (rank = D^2 forces it), reducing each Witt class to an explicit finite quadratic form (e.g. (Z_11, k^2/11), (Z_13, 2k^2/13)); SO(9)_2 is Witt-trivial. See ERRATUM_v7.md for the changelog. All computations reproducible with the included package (S unitary, S^2=C, integer Verlinde, ~1e-15).
⚠️ CORRECTION NOTICE (2026-07-12) - The boxed "Main result (Theorem)" and the abstract's "We prove an upper bound" overstate what is established: the bound depends on Assumption M1, explicitly labeled a POSTULATE (modular complexity ansatz) in this paper's own text. A contemporaneous internal multi-agent audit (3/3 independent reviewers) concluded the M1→Jacobson-EGJ bridge has "no route currently known" and recommended explicitly: "do not claim derivation." Recommend downgrading "Theorem"/"prove" to "conjectural bound under Assumption M1" in the abstract and result box.- Two papers extending Krylov complexity into the modular/horizon direction flagged as "surveillance targets" in the pre-submission novelty audit have since appeared (arXiv:2507.23739, arXiv:2602.02675); neither reproduces this paper's full composite equation, so no scoop has occurred, but the novelty margin is narrowing while this remains unsubmitted to arXiv.- This record is dormant (no update since May 2026) but not shown to be false; it sits outside the scope of the current lattice/mass-gap programme and should be read as such.``` We prove an upper bound on the rate of generalised entropy along the modular (Connes-Tomita-Takesaki) flow τ_R of a type-II crossed-product observer algebra A_R, extending the Eling-Guedens-Jacobson local Clausius schema dS = δQ/T + d_iS to the setting of Chandrasekaran-Longo-Penington-Witten and De Vuyst-Eccles-Held-Kirklin. The bound reads dS_gen[R]/dτ_R ≤ κ_R · C_k[ρ_R(τ_R)] · Θ(PH_k[δn(τ_R)]) with κ_R = 2π T_R the Bisognano-Wichmann modular surface gravity, C_k a k-design / spread complexity proxy, and Θ a topological activator on persistent homology of the dequantised classical density field δn. v6.0.16 retracts the v6.0.15 framing of M1 as a "conditional theorem" — see the framework paper (Remondière, ECI v6.0.16, DOI 10.5281/zenodo.19708889) and notes/m1_audit_2026_05_02.md. M1 retains POSTULATE status; the v8-bis math.CT companion target is Conjecture M1-C.
SUPERSEDED DUPLICATE — do not cite this record. Cite 10.5281/zenodo.21698294 instead. Both records were deposited on 2026-07-30 and contain the same paper. This one carries an earlier, superseded NOTE.md whose resource-scaling section still claimed L≈6–7 (d=36–49); the surviving record carries the corrected text (measured L=4, [[1536,6,16]]) and the toric4d_lowp.py stratified estimator behind it. The independent cross-validation scripts that were unique to this record (magic_state4-7.py) have been added to the surviving record. Flagged 2026-07-31. We simulate a 15-to-1 magic-state factory under full circuit-level depolarizing noise and ask whether a 4D toric code — which corrects errors single-shot, without repeated syndrome rounds — relieves the factory's dominant cost. It does not. Distillation output error is confirmed at 35p³ by exact enumeration and by stim, but the factory's circuit-level error floor is first order in the gate error rate and schedule-dependent, measured at 12–15·p_gate for the real [[15,1,3]] quantum Reed–Muller code with all 14 stabilizers. This is 3.5× larger than a previously reported 4.18·p_gate, which we trace to a circuit that omitted the code's 10 Z-type checks; that figure is withdrawn here. The floor originates in the final syndrome sweep: faults deposited on data qubits by the last round of gates are never checked by anything after them. Whichever check type is measured last, that sector's faults escape, so rescheduling shifts the floor between sectors and interleaving shaves roughly 20%, but cannot remove it. Reordering, redundant checks and flag qubits leave it first order; decoding weight-1 errors instead of discarding them makes it 5× worse. We construct the 4D toric code on an L⁴ torus from its cubical chain complex and verify [[486,6,9]] at L=3 (k=6 exactly, explicit weight-9 logicals in both sectors, d=4 exact at L=2 by exhaustive search), and measure a code-capacity threshold between 5% and 7% under BP+OSD. Single-shot repair using the 80 independent metachecks recovers near-perfect-syndrome performance, but only while syndrome noise stays roughly 10–30× below the data error rate: it improves logical error by 195× at q=10⁻³ and by only 4.5× at q=10⁻². Crucially, metachecks constrain the syndrome, not the data, and cannot see faults deposited directly on data qubits by the final sweep — which is precisely where the factory floor originates. Single-shot therefore buys latency, not factory fidelity. Because the factory's Cliffords act on encoded blocks, p_gate is the logical error rate of the hosting code. The correction moves the surface-code requirement from d=15 to d=17 (578 physical qubits per patch); on the 4D toric side, the smallest sufficient block is L=4, [[1536,6,16]], with [[486,6,9]] missing the fidelity demand by four to five orders of magnitude. We situate the result against Xu et al. (arXiv:2407.18490), whose parallel distillation across qLDPC patches reduces space–time cost but leaves the per-factory floor untouched. A corrections log records eight claims withdrawn or corrected during this work, including our own earlier 4.18·p_gate figure and a d=15 requirement that rested on a single error event. Section 6 states what is not established: these are code-capacity numbers for the 4D code, the distance requirement is quoted at p=3–5% rather than 10⁻³, d=L² is verified exactly only at L=2, the Λ extrapolations are long, and no end-to-end injection simulation was run. Correction, 2026-07-31. The files deposited with this record on 2026-07-30 predate a low-p re-measurement of the resource-scaling section. The originally deposited text stated that hosting the factory requires L≈6–7 (d=36–49, n≈7800–14400); that figure came from fitting a single suppression factor Λ to near-threshold points at p=3–5%. The single-Λ ansatz does not hold for this code: the measured suppression per two units of distance runs from 10.9 (d=4→9 at p=0.02) to 793 (d=9→16 at p=0.01). Re-measuring at the factory's actual operating regime with a weight-stratified estimator gives d ≥ 14.5 at p=0.02 and d ≥ 9.7 at p=0.01, so the smallest sufficient 4D block is L=4, [[1536,6,16]], not L=6–7. The L≈6–7 figure is withdrawn. These are point estimates whose upper confidence bands remain wide at the smallest p, and should be read as order-of-magnitude. The corrected text and the estimator script (toric4d_lowp.py) are not in this record's file set; replacing files requires a new version, which will carry the corrected note. Nothing in Sections 1–4 — the 35p³ distillation result, the 12–15·p_gate factory floor, the last-sweep mechanism, or the d=17 surface-code requirement — is affected.