{"id":"b02e62f9-fa53-4dd3-8be8-994f9604ac68","arxiv_id":"2412.04278","paper_version":1,"verdict":"ACCEPT","confidence":"MODERATE","novelty_score":4.0,"correctness_risk":"low","formal_verification":"none","parameter_count":2,"one_line_summary":"A Monte Carlo study shows that 3D Ising slabs with finite thickness N_z have 2D Ising critical exponents, with T_c varying smoothly from the 2D to the 3D value.","lead":"This numerical study simulates the 3D Ising magnet in a slab geometry, where two directions are large and the third stays small, and measures its critical temperature and critical exponents. It finds that a slab of finite thickness behaves like the exactly solved 2D Ising model at its phase transition, with only the critical temperature shifting smoothly toward the 3D value as thickness grows.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Hand-picked FSS plateaus (Sec. 2.6) are the key risk; a correction-to-scaling refit of the N_z=8 data would show whether the quoted 2D exponents are truly asymptotic.","rationale":"The reader's weakest assumption is exactly the one I would flag: the FSS plateaus are selected with the same data used to quote the exponents, so the systematic error from corrections to scaling is not fully captured. This is a genuine load-bearing concern because the whole universality claim hinges on those exponents being asymptotic. However, I do not think it overturns the verdict. The paper has strong internal controls: N_z=1 is the exactly solvable 2D Ising model and the same analysis recovers J_c and exponents to high precision; the 3D limit reproduces benchmark values; and the exponents for N_z=2,4,8 are all consistent with 2D values with no systematic trend in N_z. Theoretically, a finite-thickness slab with two infinite directions is expected to be in the 2D Ising universality class, so the numerical result matches a solid prior. The proposed correction-to-scaling refit would directly test whether the plateau is asymptotic, and passing it would raise confidence; failing it would identify a real problem. Since the concern is a limitation rather than a demonstrated error, the reader's ACCEPT verdict should stand unchanged.","tokens_in":5863,"tokens_out":16387,"duration_ms":173153,"concrete_test":"Re-fit the N_z=8 data for <|m|> at J_c and chi_peak using all available L with an explicit correction-to-scaling form A L^{-x}(1 + B L^{-omega}), fixing omega=2 (2D Ising leading correction) and, separately, omega=0.8 (3D Ising correction); then compare the extrapolated exponents x to beta/nu=0.125 and gamma/nu=1.75. If the asymptotic values agree with 2D and B is consistent with zero, the hand-selected Lmin is not biasing the result. If the extrapolation moves toward 3D values or depends strongly on the assumed omega, the plateau in Fig. 5 is not asymptotic and the central claim needs stronger evidence.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central claim—every fixed-N_z slab is in the 2D Ising universality class—rests on the exponents in Table 1, obtained by fitting Eq. (3) to data with L >= Lmin, where Lmin is selected by excluding small lattices until the estimator 'does not change significantly any more' (Sec. 2.6). This selection rule is data-dependent: the plateaus in Fig. 5 may reflect a slowly decaying correction-to-scaling term or statistical noise rather than the true asymptotic 2D regime. If, over the accessible range (e.g., L=256–2048 for N_z=8), the effective exponents are still drifting toward 2D from the 3D side, the jackknife errors on the plateau values understate the systematic uncertainty, and the conclusion that the exponents are exactly 2D would not be established. The N_z=1 and 3D control runs validate the analysis pipeline, but they do not eliminate this risk, because the finite-thickness crossover could create a spurious plateau at intermediate N_z.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"This paper reports a numerical Monte Carlo study of the ferromagnetic Ising model on L×L×N_z lattices with periodic boundary conditions, covering the ordinary 3D case (N_z = L, L = 24–256) and the dimensionally reduced cases with fixed N_z = 1, 2, 4, 8 and lateral sizes up to L = 2048 for N_z = 8. The simulations use a combined Metropolis and Wolff cluster update, histogram reweighting, and delete-d jackknife error estimates with d ≫ 2τ_int + 1. The critical coupling is extracted from Binder-cumulant crossings, and the exponents β/ν, γ/ν, and ν are extracted from finite-size scaling fits of Eq. (3) with small lattices excluded stepwise until a plateau in Lmin is reached. The authors find that T_c(N_z) varies smoothly from the exact 2D value at N_z = 1 to the 3D value as N_z grows, and that for each fixed N_z the measured exponents agree within errors with the 2D Ising values (β/ν = 0.125, γ/ν = 1.75, ν = 1), with an effective dimension d_eff = 2. They conclude that every finite-thickness slab of the 3D Ising model remains in the 2D Ising universality class.","tokens_in":6077,"tokens_out":17647,"duration_ms":164818,"significance":"The result, if correct, is a clean numerical confirmation of the expected dimensional crossover for magnetic thin films: for a slab of fixed thickness with a thermodynamic limit in the other two directions, the finite thickness is a non-universal parameter that shifts T_c but does not change the universality class, here the 2D Ising class. The paper's strengths are its methodological care and its controlled benchmarks: the N_z = 1 results reproduce the exact Onsager value of T_c and the 2D exponents, and the 3D results agree with the high-precision values of Ferrenberg et al. [11]; the lateral volumes used for N_z = 8 (up to L = 2048) are large for this kind of study; and the jackknife blocking with d ≫ 2τ_int + 1 is sound. The T_c(N_z) curve provides a useful benchmark for other slab studies, and the universality claim is a falsifiable prediction that is here tested for N_z = 1, 2, 4, 8. The main caveat is that the exponent evidence rests on hand-selected Lmin plateaus without an explicit correction-to-scaling analysis, so the statistical errors in Table 1 do not yet include the relevant systematic uncertainty.","major_comments":[{"comment":"The central universality claim rests on the exponents in Table 1, which are obtained by fitting Eq. (3) to data with L ≥ Lmin, where Lmin is chosen by the data-dependent criterion of Sec. 2.6, namely excluding small lattices until the estimator \"does not change significantly any more\". A slowly decaying correction-to-scaling term could sustain such a plateau over the accessible range (for N_z = 8, L = 256–2048) while the true asymptotic value is still approached from the 3D side, and the reported jackknife errors contain no component for the Lmin choice. Please add a quantitative correction-to-scaling analysis, for instance fits of the form χ_max = A L^{γ/ν}(1 + B L^{-ω}), ⟨|m|⟩|_{J̃_c} = A' L^{-β/ν}(1 + B' L^{-ω'}), and ∂U4/∂J̃|_{J̃_c} = A'' L^{1/ν}(1 + B'' L^{-ω''}) over the full L range, or at least a documented plateau criterion (e.g., χ²/dof ≤ 1 for L ≥ Lmin and stability of the central value within a fraction of the statistical error across successive Lmin steps), together with an explicit systematic error added to Table 1. Without such a step, the statement that the slab exponents are exactly the 2D Ising values is asserted at a precision that the current analysis does not fully support.","section":"Sec. 2.6, Fig. 5, Table 1"},{"comment":"The T_c estimators are likewise formed from hand-picked crossing plateaus: for N_z = 2 the weighted average uses crossings with L1 ≥ 128 and for N_z = 4 those with L1 ≥ 512, and the quoted uncertainty (e.g., 0.23602775(15) for N_z = 4) is the jackknife error of that weighted average only. Since Fig. 3 shows a clear monotone drift of the crossings below the chosen onset, the reported error understates the uncertainty if the onset is misidentified. Please report the sensitivity of J̃_c to the plateau onset (for instance, the values and errors obtained with several alternative onsets around the chosen one) and either incorporate the residual spread into the quoted errors or justify the chosen onset with a quantitative criterion.","section":"Sec. 3.1, Fig. 3"}],"minor_comments":[{"comment":"The abstract and introduction say that β, γ, and ν are determined, but Table 1 lists only the ratios β/ν, γ/ν, and ν; please rephrase to state that the ratios are determined directly, with β and γ following from multiplication by ν.","section":"Abstract and Sec. 1"},{"comment":"The sentence \"For any finite N_z our critical exponents suggest that the model is still in the 2D Ising model universality class\" extrapolates beyond the simulated values N_z = 1, 2, 4, 8; either restrict the claim to the values studied or add a supporting argument (for instance, finiteness of the transfer matrix in the z-direction) for why the universality class is independent of N_z.","section":"Sec. 4 (Conclusions)"},{"comment":"The axis labels for the inverse thickness are typeset in a broken form (\"N □1 z\" and \"N−1 z\"); please ensure that N_z^{-1} is rendered with a proper superscript in the final version.","section":"Figs. 3 and 6"},{"comment":"The statement that O(10^6–10^8) measurements are performed is very broad; a short table listing the number of measurements and the integrated autocorrelation time τ_int for each (N_z, L) ensemble would substantially improve reproducibility.","section":"Sec. 2.1"},{"comment":"The sentence \"The latter has been investigated in Refs. [8–11]\" groups two thin-film papers [9,10] with the 3D bulk studies [8,11]; it would be clearer to cite Ferrenberg et al. [11] for the 3D bulk value and to place Refs. [9,10] in the dimensional-reduction context where they belong.","section":"Refs. [8-11]"},{"comment":"The caption lists the L1 values in a compressed typeset form; please list the full sequence of L1 values for each panel so the reader can identify the plateau onset directly from the caption.","section":"Fig. 3 caption"}],"recommendation":"major_revision","confidential_remarks":"To the editor: the paper is a PoS proceedings contribution whose central claim is physically expected rather than controversial, and the authors are appropriately hedged in their wording. My main reservation is that the headline numbers in Table 1 carry only statistical errors while the dominant systematic (the Lmin plateau choice) is handled by a heuristic, and I would like to see that addressed before publication. The required changes are local and, in my view, likely to leave the conclusions intact. No issues with citation fairness or novelty disclosure; the relevant thin-film literature is cited."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Quick take: this is a careful numerical study of 3D Ising slabs with fixed thickness N_z, and the new thing is precision: beta/nu, gamma/nu, and nu for N_z=2,4,8, plus a T_c curve that interpolates smoothly between the 2D and 3D values. The conclusion — every finite-N_z slab is in the 2D Ising universality class — is expected from thin-film physics and dimensional crossover arguments, so the surprise value is modest. But the numbers are solid and more precise than anything I know for these thicknesses, and they'll be useful benchmarks for effective 2D models and thin-film simulations.\n\nThe paper does things right. The update scheme (Metropolis + Wolff), histogram reweighting, and jackknife with d >> 2τ+1 are standard. The N_z=1 and 3D control runs agree with exact and benchmark results, which validates the pipeline. The Lmin-dependence plots in Fig. 5 show plateaus within errors, and the quoted exponents sit right on the 2D values.\n\nThe real soft spot is the Lmin selection rule in Sec. 2.6: the authors exclude small lattices until the estimator \"does not change significantly any more.\" That's data-dependent. The plateaus in Fig. 5 could in principle be a slowly decaying correction-to-scaling term masquerading as the asymptotic regime, and the jackknife errors on the plateau values don't capture that systematic uncertainty. The stress-test worry about a spurious plateau at intermediate N_z is worth taking seriously, but the controls at N_z=1 and the 3D limit, plus the fact that the exponents land on the 2D values for all four N_z, make me think the conclusion is right. A correction-to-scaling fit on the N_z=8 data would tighten it further, but its absence doesn't sink the paper.\n\nMinor complaints: no shipped code or data tables, which keeps confidence at moderate. The paper is a proceedings-style contribution, so the brevity is fine.\n\nBottom line: for anyone working on thin films, dimensional crossover, or validating effective 2D descriptions, this is a useful benchmark. It deserves a serious referee, and I'd be comfortable accepting after a request for a clearer statement of the Lmin systematics. Not desk-reject material.","headline":"Precise, careful Monte Carlo benchmarks for fixed-thickness 3D Ising slabs; the 2D universality conclusion is expected but the numbers are worth having, with Lmin systematics the one caveat.","tokens_in":6613,"tokens_out":1751,"would_cite":true,"duration_ms":16539,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"deepseek-v4-flash","headline":"For any fixed thickness $N_z$, the dimensionally reduced 3D Ising model shows the 2D Ising critical exponents $\\beta/\\nu=1/8$, $\\gamma/\\nu=7/4$, $\\nu=1$, while the critical coupling moves smoothly from the exact 2D value to the 3D value…","keywords":["three-dimensional Ising model","dimensional reduction","critical exponents","universality class","finite-size scaling","histogram reweighting","Binder cumulant","Monte Carlo simulation"],"falsifier":"Simulate $L\\times L\\times N_z$ for $N_z=4$ or 8 with $L=4096$ and $L=8192$; if $\\gamma/\\nu$ and $\\nu$ drift away from 1.75 and 1 toward the 3D values (about 1.963 and 0.629) as $L$ grows past the current plateaus, the claimed 2D universality is false, whereas if they stay flat the claim is supported.","tokens_in":5648,"feed_emoji":"🧲","tokens_out":6728,"duration_ms":55937,"temperature":0.7,"pith_summary":"The paper asks whether the three-dimensional Ising model, when squeezed to a fixed finite thickness in one direction, still looks two-dimensional at its phase transition. Simulating lattices $L\\times L\\times N_z$ with $N_z=1,2,4,8$ and $L$ up to 2048, the authors extract the critical coupling and the ratios $\\beta/\\nu$, $\\gamma/\\nu$ and $\\nu$ by finite-size scaling and histogram reweighting. They find a smooth curve of critical couplings connecting the exact 2D value to the 3D value, while the exponents at every fixed $N_z$ match the 2D Ising values. The conclusion is that any finite-thickness slab of the 3D Ising model belongs to the 2D Ising universality class, with the critical temperature shifting smoothly with thickness.","feed_headline":"Every finite-thickness 3D Ising slab still shows 2D exponents","feed_subtitle":"Simulations trace a smooth critical-coupling curve from the exact 2D value to 3D, with 2D exponents all along.","key_machinery":"The central objects are the finite-size scaling relations of Eq. (3): at the critical coupling, $\\langle|m|\\rangle \\propto L^{-\\beta/\\nu}$, the peak susceptibility $\\max \\chi \\propto L^{\\gamma/\\nu}$, and $dU_4/d\\tilde J \\propto L^{1/\\nu}$. These ratios are extracted from Binder cumulant crossings, which locate $\\tilde J_c$, and from susceptibility peaks located by histogram reweighting, with jackknife errors and a plateau criterion in $L_{\\min}$ to suppress corrections to scaling; the effective dimension $d_{\\rm eff}=(2\\beta+\\gamma)/\\nu$ then diagnoses the universality class.","core_discovery":"For a fixed finite $N_z$, the dimensionally reduced 3D Ising model has the same critical exponents as the 2D Ising model, regardless of $N_z$: $\\beta/\\nu \\approx 0.125$, $\\gamma/\\nu \\approx 1.75$, $\\nu \\approx 1$, giving an effective dimension $d_{\\rm eff}=(2\\beta+\\gamma)/\\nu \\approx 2.000$ for all simulated thicknesses. The critical coupling $\\tilde J_c$ moves continuously from $0.44068679\\ldots$ at $N_z=1$ to the 3D value $0.22165494(49)$ as $N_z$ grows, and the paper's 3D results agree with the best existing 3D Ising estimates. The paper's central claim is that the 2D universality class persists for every finite $N_z$.","pith_inferences":["One could test whether the same dimensional-reduction pattern holds in other universality classes, for example the 3D O($N$) models, where fixed-thickness slabs might or might not stay in the lower-dimensional class.","The plateau-in-$L_{\\min}$ criterion could hide corrections to scaling that mimic 2D exponents over the simulated range; an analysis with explicit subleading-exponent fits would settle whether the 2D values are asymptotic or a finite-range accident.","If the claim holds, the crossover between 2D and 3D criticality is not a gradual change in exponents but a sharp dimensional reduction at any finite thickness, which has practical implications for finite-size extrapolations in lattice simulations.","A scaling-collapse analysis of the Binder cumulants at fixed $N_z$, varying both $L$ and $N_z$, would provide an independent check of the claimed universality class."],"forward_implications":["Any finite-thickness slab of the 3D Ising model can be treated as a 2D Ising system at criticality: only the critical temperature is shifted, not the exponents.","The measured $\\tilde J_c(N_z)$ curve provides a precise interpolation between the exact 2D and the 3D critical couplings, which a future dimensional-crossover formula should reproduce.","The effective dimension $d_{\\rm eff}\\approx 2.000$ at all simulated $N_z$ means the critical behavior of a slab is genuinely two-dimensional even when the slab is eight layers thick.","The same finite-size scaling analysis at fixed $N_z$ supplies reference values that other methods, such as tensor networks or thin-film experiments, can test against."],"supporting_citations":[{"why":"Cited as the exact 2D Ising solution; supplies the analytic critical coupling and the $\\beta/\\nu=1/8$, $\\gamma/\\nu=7/4$, $\\nu=1$ values used as the comparison baseline.","marker":"[1]"},{"why":"Introduces the fourth-order Binder cumulant $U_4$ whose crossing technique locates the critical coupling $\\tilde J_c$.","marker":"[2]"},{"why":"Supplies the finite-size scaling relations in Eq. (3) that connect lattice size $L$ to magnetization, susceptibility and cumulant derivative, and hence to the critical exponents.","marker":"[3-5]"},{"why":"Provides histogram reweighting, which lets the analysis vary $\\tilde J$ continuously to locate susceptibility peaks and cumulant crossings.","marker":"[6,7]"},{"why":"Earlier dimensional-crossover study of the 3D gauge Ising model; the paper's $\\tilde J_c$ and $\\nu$ at fixed $N_z$ are compatible with (and more precise than) its results.","marker":"[8]"},{"why":"High-precision 3D Ising reference; the paper's 3D values for $\\tilde J_c$ and the exponents are checked against it.","marker":"[11]"}],"fun_headline_variants":["Thin 3D Ising slabs keep 2D critical exponents","Finite-thickness Ising slabs mimic 2D universality","Slab thickness doesn't change Ising 2D exponents","3D Ising slab criticality stays two-dimensional","Reduced-dimension Ising model stays 2D-like"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The finite-size scaling relations are assumed to hold, with negligible subleading corrections, for lattice sizes above a hand-chosen cutoff $L_{\\min}$, which is selected by dropping small lattices until the exponent estimates stop changing; if those plateaus are not the true asymptotic regime, the 2D-universality conclusion would not follow.","fun_headline_variants_meta":{"raw":{"variants":["Thin 3D Ising slabs keep 2D critical exponents","Finite-thickness Ising slabs mimic 2D universality","Slab thickness doesn't change Ising 2D exponents","3D Ising slab criticality stays two-dimensional","Reduced-dimension Ising model stays 2D-like"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000472,"raw_usage":{"total_tokens":2337,"prompt_tokens":924,"completion_tokens":1413,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":540,"completion_tokens_details":{"reasoning_tokens":1326}},"tokens_in":540,"tokens_out":1413,"duration_ms":10656,"temperature":1.0,"reasoning_tokens":1326,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-11T21:34:29.363702+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Simulate $L\\times L\\times N_z$ for $N_z=4$ or 8 with $L=4096$ and $L=8192$; if $\\gamma/\\nu$ and $\\nu$ drift away from 1.75 and 1 toward the 3D values (about 1.963 and 0.629) as $L$ grows past the current plateaus, the claimed 2D universality is false, whereas if they stay flat the claim is supported.","supporting_citations":[{"cited_title":"Ising,Beitrag zur Theorie des Ferromagnetismus, Z","cited_arxiv_id":null,"evidence_quote":"Cited as the exact 2D Ising solution; supplies the analytic critical coupling and the $\\beta/\\nu=1/8$, $\\gamma/\\nu=7/4$, $\\nu=1$ values used as the comparison baseline."},{"cited_title":"Binder,Finite size scaling analysis of ising model block distribution functions, Z","cited_arxiv_id":null,"evidence_quote":"Introduces the fourth-order Binder cumulant $U_4$ whose crossing technique locates the critical coupling $\\tilde J_c$."},{"cited_title":"Deconfinement transition and dimensional cross-over in the 3D gauge Ising model","cited_arxiv_id":"hep-lat/9511015","evidence_quote":"Earlier dimensional-crossover study of the 3D gauge Ising model; the paper's $\\tilde J_c$ and $\\nu$ at fixed $N_z$ are compatible with (and more precise than) its results."}],"review_version":1}