{"id":"174fb2a8-0247-4fb6-888a-7350db35c86d","arxiv_id":"2412.13407","paper_version":2,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":3,"one_line_summary":"Non-perturbative lattice calculations determine the BMN deconfinement temperature from the perturbative to the holographic regime, with evidence that the transition becomes continuous at weak coupling.","lead":"Lattice simulations of the BMN matrix model map the deconfinement temperature across three orders of magnitude in coupling, interpolating between perturbation theory and a supergravity prediction. The results also suggest the phase transition, first order at strong coupling, may become continuous at weaker couplings.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Two-lattice-spacing averaging without a true Nτ→∞ extrapolation leaves an unquantified discretization bias in (T/µ)_crit, the central quantity of the paper's interpolation claim.","rationale":"The reader's weakest assumption identifies the same load-bearing concern: two lattice spacings are averaged rather than extrapolated, so continuum-limit control of (T/µ)_crit is missing. This is the right focal point because the paper's headline result is the quantitative phase diagram, and an unquantified discretization bias could compromise both the weak-coupling agreement with perturbation theory and the strong-coupling approach to supergravity. The order-of-transition analysis is also fragile, but it is secondary to the interpolation claim and the reader has already flagged it; our stress-test concentrates on the factor most likely to change the central conclusion. We therefore agree with the reader's conditional-accept recommendation and do not propose a different verdict.","tokens_in":19341,"tokens_out":8825,"duration_ms":78029,"concrete_test":"Reanalyze the open data to obtain (T/µ)_crit for each Nτ separately using multi-histogram Ferrenberg–Swendsen reweighting of the Polyakov-loop susceptibility, and compare the Nτ = 8/16 and 16/24 pairs with full jackknife errors. Then perform a genuine continuum extrapolation in 1/Nτ^2, or at least assign a systematic error from the difference between the two spacings. To settle the concern, run or locate additional ensembles at Nτ = 32 for g = 10^-5 and g = 0.01 (e.g., exploiting the SU(4) large-Nτ runs from Sec. III if Polyakov-loop data exist there) and check whether the resulting (T/µ)_crit moves by more than the quoted uncertainties. If the extrapolated or Nτ = 32 values agree with the reported averages, the interpolation claim is secure; if they shift, the central phase diagram requires revision.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central claim is a quantitative interpolation of (T/µ)_crit between weak-coupling perturbation theory and strong-coupling supergravity. Each (T/µ)_crit is determined from only two lattice spacings, then 'average[d]... by fitting to a constant' (Sec. IV), while the Fig. 6 caption labels these results as 'Nτ→∞ extrapolations'. No genuine continuum extrapolation is performed, and the Sec. III checks (SO(6) breaking, Pfaffian phase) target different observables and do not bound the discretization error of the Polyakov-loop transition temperature itself. At the weak-coupling points, the lattice spacing is not small: aµ = 1/(Nτ T/µ) ≈ 1.6 for Nτ = 8 and ≈ 0.8 for Nτ = 16, so O(a^2) artifacts are not obviously negligible. The sigmoid fits used for the central values have large χ²/d.o.f. (up to 54), indicating underestimated uncertainties that make the two-point 'agreement' less informative. If residual Nτ effects shift the true continuum values by more than the quoted errors, the claimed agreement with NNLO perturbation theory at g ≤ 10^-4 and the approach to the supergravity value at g ≥ 10^-3 could be weakened or altered, directly undermining the central claim.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The manuscript reports lattice Monte Carlo determinations of the critical deconfinement temperature of the BMN matrix model for couplings g = 10^-5 to 10^-2, using a simple one-dimensional lattice action with RHMC and phase-quenched Pfaffian. For each of four couplings and N = 8, 12, 16, the authors scan T/μ, locate susceptibility peaks in the Polyakov loop, fit the Polyakov loop to a sigmoid, and then average the two lattice sizes per {g,N}. They compare the resulting phase diagram with NNLO perturbation theory and the strong-coupling supergravity prediction, and use the growth of the susceptibility peak with N^2 to infer the order of the transition.","tokens_in":19623,"tokens_out":6809,"duration_ms":63516,"significance":"If the central results survive revision, they would constitute a useful non-perturbative bridge between the perturbative and holographic regimes of the BMN model, extending earlier studies with a different lattice action and an open data release. The consistency across N and Nτ, the explicit checks of the Pfaffian phase, and the reproducible workflow are genuine strengths. However, as discussed below, the quantitative interpolation claim and the weak-coupling order claim currently rely on unquantified systematics, so the manuscript needs substantive revision before the advertised conclusions are fully supported.","major_comments":[{"comment":"The points labelled 'Nτ → ∞ extrapolations' in Fig. 6 are obtained by fitting a constant to exactly two lattice sizes (Nτ = 8 and 16 or 16 and 24). This is an average, not a continuum extrapolation: no scaling form is used, no third lattice size constrains the trend, and the agreement between two points does not bound residual discretization effects. At the weak-coupling points aμ ≈ 1/(Nτ T/μ) is ≈ 1.6 for Nτ = 8 and ≈ 0.8 for Nτ = 16, so the lattice spacing is not parametrically small; the Sec. III checks address the SO(6)-breaking ratio and the Pfaffian phase, not the convergence of the Polyakov-loop transition temperature itself. Since the abstract's central claim is a quantitative interpolation between Eq. (10) and Eq. (11), an unquantified discretization bias in (T/μ)_crit directly affects that claim. I ask the authors to either perform a genuine Nτ → ∞ extrapolation (even a simple 1/Nτ^2 ansatz with additional Nτ values) or re-label the results as averages of two lattice sizes and include an estimated discretization systematic.","section":"Sec. IV / Fig. 6"},{"comment":"The sigmoid fits used for the central values have χ²/d.o.f. as large as 54 (e.g., g = 0.001, N = 12, Nτ = 24), and the authors correctly note that the quoted fit uncertainties are underestimates. Switching to the coarser susceptibility-peak uncertainties changes the error bars but not the central values; if the sigmoid ansatz is disfavoured by the data, the fit central values themselves can be biased. The T/μ grid spacing is coarse, so the 'conservative' susceptibility-peak uncertainties may not span this bias. I ask for a systematic account of fit-model and fit-range variation, or a reweighting-based interpolation, before using these central values in the comparison with perturbation theory.","section":"Table I / Eq. (24)"},{"comment":"The claim that the transition is continuous for g ≲ 10^-4 is based on two lattice sizes and two gauge groups per coupling, with exponent estimates b = 0.85 → 0.61 (g = 10^-4) and 0.71 → 0.59 (g = 10^-5) obtained without uncertainties and from two-point power-law fits. There is no Nτ → ∞ extrapolation of b and no demonstration that the trend is not a finite-N or finite-Nτ artifact. This is a secondary but advertised conclusion ('appears to be continuous for weaker couplings' in the abstract). Please either add a proper finite-size scaling analysis with more N values and error propagation, or explicitly downgrade this statement to a qualitative observation.","section":"Sec. IV, final paragraph / Fig. 8"}],"minor_comments":[{"comment":"The notation '26 · 5/34' and the corresponding expression for C_NNLO appear garbled by lost superscripts; please typeset as 2^6 · 5 / 3^4 and the analogous NNLO expression.","section":"Eq. (10)"},{"comment":"The last x-axis tick reads '0 .035', which looks like a formatting artifact; please fix the axis labeling.","section":"Fig. 1"},{"comment":"Since the text repeatedly refers to the averaged critical temperatures, consider reporting the averaged value and a combined systematic uncertainty in Table I, rather than only the individual Nτ entries.","section":"Table I / Sec. IV"},{"comment":"When comparing the right-most g = 0.001 point in Fig. 3 with Fig. 2, the sentence says 'apart from the different gauge group' but does not state the comparison values; please spell out the two numbers being compared.","section":"Sec. III, Fig. 3"},{"comment":"Please clarify explicitly that the sigmoid parameter D is the inflection point of the functional form as written, since this is the identification used to define (T/μ)_crit.","section":"Eq. (24)"}],"recommendation":"major_revision","confidential_remarks":"To the editor: the manuscript is honest about many of its limitations, including the large sigmoid χ²/d.o.f. and the absence of true continuum extrapolations, but the abstract and Fig. 6 nevertheless overstate the continuum-limit status of the results. The open data release makes the requested re-analysis feasible, and I see no grounds for rejection. However, the 'Nτ → ∞ extrapolation' labeling and the weak-coupling order-of-transition claim need substantive revision before publication."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Quick take: this is a solid lattice Monte Carlo study of the BMN matrix model that delivers the advertised interpolation of the deconfinement temperature between weak-coupling perturbation theory and the strong-coupling supergravity value. Worth engaging, but read the continuum-limit claims carefully. The text is honest about averaging two lattice sizes, while the Fig. 6 caption calls those results \"Nτ→∞ extrapolations\". That is an overstatement, not a fraud — the pairs agree within errors and the main phase diagram is likely right.\n\nWhat is new: extending previous scans of (T/µ)_crit to weaker couplings g = 1e-5 and 1e-4 with a different scan strategy, plus explicit checks of the Pfaffian phase and SO(6) breaking. The weak-coupling points sit on the NNLO perturbative curve and the stronger-coupling points approach the supergravity value, so the central claim is plausible. The open data release and public software make this reproducible work, which should earn real credit.\n\nSoft spots, in proportion. First, no genuine continuum extrapolation: two lattice spacings per parameter set are averaged by a constant fit, not extrapolated. Residual a^2 effects are unquantified, and at weak coupling the coarser spacing has aµ ≈ 1.6, which is not small. This does not obviously invalidate the phase diagram because the paired results are mutually consistent, but the uncertainties on (T/µ)_crit are understated and the caption needs fixing. Second, the continuous-transition claim for g ≲ 1e-4 rests on only two gauge groups and two lattice sizes, with critical exponents quoted without uncertainties. The authors say \"neglecting uncertainties\" and frame it as suggestive, which is honest, but a referee should push for error bars or a softened abstract. Third, the sigmoid fits have large χ²/d.o.f.; using the susceptibility-peak uncertainties instead is a sensible heuristic, but the reported errors are not fully controlled.\n\nThe Pfaffian phase handling is fair: they show it is small in the relevant regime and clearly state the phase-quenched approximation. Not propagating it as a systematic is defensible given those checks.\n\nOverall: the central interpolation is a useful non-perturbative benchmark, but the order-of-transition conclusion is not yet solid. Send it to a serious referee; expect heavy revision on the continuum-limit language and the scaling analysis. I would cite the dataset if I were doing related numerics.","headline":"A competent, reproducible lattice study that plausibly interpolates between perturbative and supergravity predictions, but the continuum-limit language overstates what two lattice spacings can support and the weak-coupling continuous-transition claim is still speculative.","tokens_in":20181,"tokens_out":2293,"would_cite":true,"duration_ms":28841,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"deepseek-v4-flash","headline":"This paper uses lattice Monte Carlo to compute the finite-temperature phase diagram of the BMN matrix model, a maximally supersymmetric quantum-mechanical system with a well-defined thermal partition function, and finds that the…","keywords":["BMN matrix model","deconfinement transition","Polyakov loop","lattice supersymmetry","gauge/gravity duality","supergravity","Monte Carlo simulation","phase diagram"],"falsifier":"A direct calculation at g = 0.01, N = 12 with Nτ = 32 or 48, comparing (T/µ)_crit to the quoted average 0.100, would settle whether the constant-average continuum-limit assumption holds; alternatively, a high-statistics measurement of the susceptibility exponent b for g = 10⁻⁴ with N = 16 would settle whether it continues to decrease toward the continuous-transition value.","tokens_in":19110,"feed_emoji":"🌡️","tokens_out":7455,"duration_ms":60718,"temperature":0.7,"pith_summary":"The paper uses lattice Monte Carlo to map the deconfinement transition of the BMN matrix model, a maximally supersymmetric quantum-mechanical theory with a well-defined thermal partition function. It determines the critical temperature (T/µ)_crit as a function of the dimensionless coupling g = λ/µ³, spanning three orders of magnitude. The central result is that the lattice results interpolate between the weak-coupling perturbative prediction (T/µ ≈ 1/(12 ln 3) ≈ 0.076, with known NNLO corrections) and the strong-coupling large-N supergravity value (T/µ ≈ 0.106). A secondary result is that the transition is first order for strong couplings but appears to become continuous for g ≲ 10⁻⁴. If correct, this provides a non-perturbative bridge between two analytic limits within a single holographic system.","feed_headline":"Lattice curve spans weak and strong coupling in BMN model","feed_subtitle":"Matches perturbation theory at one end and supergravity at the other, hinting at a change in transition order.","key_machinery":"The analysis is carried by the Polyakov loop, the traced Wilson line around the thermal circle, which is an order parameter for spontaneous breaking of the Z_N center symmetry: it vanishes in the confined phase and is nonzero in the deconfined phase. The transition temperature is extracted both from the peak of the Polyakov-loop susceptibility and from a four-parameter sigmoid fit Σ = A − B/(1+exp[C(T/µ−D)]) to the Polyakov loop itself, with D locating the inflection point. To determine the order of the transition, the authors fit the growth of the susceptibility peak with N², the number of degrees of freedom, to a power law χ_max = C $N^{{2b}}$, comparing b to the first-order value b = 1.","core_discovery":"For each fixed coupling g, the authors scan in temperature T/µ and measure the Polyakov loop, locating the deconfinement transition from the peak of its susceptibility and from a sigmoid fit to the Polyakov loop itself. Comparing these determinations across couplings, they find (T/µ)_crit ≈ 0.076 at g ≤ 10⁻⁴, in agreement with NNLO perturbation theory, rising monotonically with g and approaching the g→∞ supergravity prediction (T/µ)_crit ≈ 0.106 at g = 0.01. The susceptibility-peak exponent b from scaling χ_max ∝ $N^{{2b}}$ is consistent with first-order behavior b = 1 for g ≳ 0.001, while decreasing toward b ≈ 0.59–0.61 as Nτ increases for g ≤ 10⁻⁴, which the authors read as evidence that the transition becomes continuous at weak coupling.","pith_inferences":["If the weak-coupling transition is truly continuous, the first-order line at strong coupling must terminate at a critical endpoint g⋆ somewhere in 10⁻⁴ ≲ g ≲ 10⁻³; a dedicated scan across that interval with N ≥ 16 and Nτ ≥ 24 could map it directly.","The near-N-independence of (T/µ)_crit for N = 8–16 hints that the large-N limit may already be effectively saturated at these couplings; a direct N = 32 run at g = 0.01 would test whether finite-N corrections are truly absent.","Because only two lattice sizes were averaged for each {g,N} pair, the continuum-limit assumption could be checked by a single Nτ = 32 or 48 run at fixed g and N; if the result shifts beyond the quoted error, the reported phase diagram would need to be revised."],"forward_implications":["The lattice results provide a continuous curve for (T/µ)_crit(g) that interpolates between the weak-coupling NNLO prediction and the strong-coupling supergravity limit, offering a direct target for dual-supergravity calculations of the coupling dependence.","The agreement at g ≤ 10⁻⁴ confirms that the NNLO perturbative expansion with expansion parameter 27g is accurate up to 27g ≈ 0.003, and that the phase-quenched RHMC calculations are reliable in this regime.","At strong couplings g ≥ 0.001, the susceptibility exponent b ≈ 0.87–0.91 with a trend toward 1 as Nτ increases is consistent with the first-order transition predicted in the large-N, strong-coupling limit.","The decrease of the susceptibility exponent toward b ≈ 0.59–0.61 for g ≤ 10⁻⁴ suggests the transition becomes continuous at weak coupling, implying a critical endpoint separating the strong-coupling first-order line.","The verification that Pfaffian phase fluctuations vanish in the continuum limit removes a potential sign problem for the lattice implementation of the BMN model, supporting further simulations at larger N and Nτ."],"supporting_citations":[{"why":"Supplies the strong-coupling supergravity limit (T/µ)_crit ≈ 0.106 that the lattice results approach.","marker":"[51]"},{"why":"Provides the O(g) and O(g²) perturbative corrections to the weak-coupling critical temperature used for comparison.","marker":"[54]"},{"why":"Gives the perturbative computation of the phase transition and the effective-action argument that it remains first order at weak coupling.","marker":"[55]"},{"why":"Earlier lattice study of the BMN model phase diagram whose results are compared and extended.","marker":"[27]"},{"why":"Earlier lattice study reaching stronger couplings, providing the comparison for strong-coupling behavior and finite-N effects.","marker":"[29]"},{"why":"Establishes the lattice formulation approach for maximally supersymmetric matrix models and the absence of fine-tuning.","marker":"[3]"},{"why":"Earlier lattice study of the BMN model using the same discretization and providing a critical-coupling comparison.","marker":"[11]"},{"why":"Preliminary conference proceedings introducing the sigmoid ansatz and the preliminary BMN results that this work finalizes.","marker":"[28]"}],"fun_headline_variants":["BMN transition: first-order at strong, continuous at weak coupling","Lattice BMN: deconfinement order shifts with coupling strength","BMN deconfinement: order changes as coupling weakens","BMN matrix model: transition order hinges on coupling"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The central claims assume that the two lattice sizes available per coupling and gauge group are sufficient to treat the average as the continuum limit, with no Nτ→∞ extrapolation and no quantified discretization error.","fun_headline_variants_meta":{"raw":{"variants":["BMN transition: first-order at strong, continuous at weak coupling","Lattice BMN: deconfinement order shifts with coupling strength","BMN deconfinement: order changes as coupling weakens","BMN matrix model: transition order hinges on coupling"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000208,"raw_usage":{"total_tokens":1368,"prompt_tokens":875,"completion_tokens":493,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":491,"completion_tokens_details":{"reasoning_tokens":420}},"tokens_in":491,"tokens_out":493,"duration_ms":5051,"temperature":1.0,"reasoning_tokens":420,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-11T13:09:47.606013+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"A direct calculation at g = 0.01, N = 12 with Nτ = 32 or 48, comparing (T/µ)_crit to the quoted average 0.100, would settle whether the constant-average continuum-limit assumption holds; alternatively, a high-statistics measurement of the susceptibility exponent b for g = 10⁻⁴ with N = 16 would settle whether it continues to decrease toward the continuous-transition value.","supporting_citations":[{"cited_title":"A non-perturbative formulation of N=4 super Yang-Mills theory based on the large-N reduction","cited_arxiv_id":"1106.5590","evidence_quote":"Supplies the strong-coupling supergravity limit (T/µ)_crit ≈ 0.106 that the lattice results approach."},{"cited_title":"Two-Loop Partition Function in the Planar Plane-Wave Matrix Model","cited_arxiv_id":"hep-th/0409178","evidence_quote":"Gives the perturbative computation of the phase transition and the effective-action argument that it remains first order at weak coupling."},{"cited_title":"Thermal phase structure of a supersymmetric matrix model","cited_arxiv_id":"2003.01298","evidence_quote":"Earlier lattice study reaching stronger couplings, providing the comparison for strong-coupling behavior and finite-N effects."},{"cited_title":"Non-lattice simulation of supersymmetric gauge theories as a probe to quantum black holes and strings","cited_arxiv_id":"0912.0327","evidence_quote":"Earlier lattice study of the BMN model using the same discretization and providing a critical-coupling comparison."}],"review_version":1}