{"id":"916fd707-0299-4842-9d76-703fd55a7498","arxiv_id":"2501.15494","paper_version":2,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":5.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":5,"one_line_summary":"A new large-volume lattice QCD run with Möbius domain wall fermions finds the Nf=3 chiral transition at T=121(2) MeV is consistent with a smooth crossover near m_f^MS(2 GeV)=4 MeV.","lead":"Lattice QCD simulations with three degenerate quark flavors at a temperature of about 121 MeV find that the transition behaves as a smooth crossover, not a sharp phase transition, at a quark mass near 4 MeV. The result adds a new large 48^3 lattice and a check of residual chiral symmetry breaking to earlier smaller-volume runs, supporting the emerging consensus from staggered and Wilson fermion studies.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Central claim of a crossover rests on a single lattice spacing; no continuum extrapolation is performed, so the result may be a discretization artifact.","rationale":"The reader's weakest assumption identifies the single-lattice-spacing issue, and I agree that this is the most load-bearing concern. The central claim is a statement about the physical QCD transition, yet all finite-temperature data are at a=0.1361 fm. Because the transition order is known to be sensitive to the lattice action and spacing (as the paper's own Introduction documents), there is no reason to assume that the observed crossover survives the continuum limit. I considered other potential weaknesses—the qualitative nature of the finite-size scaling (errors on B4 are not given) and the model-dependent subtraction of the UV divergence—but these could at most weaken the evidence at the given spacing; they would not invalidate the conclusion if the continuum limit also shows a crossover. The single-spacing issue is decisive: if a finer lattice reveals a first-order or Z(2) transition, the paper's conclusion is wrong even if all present data are as reported. The proposed test is to run the same observables at a finer spacing with comparable temperature and quark mass, using at least two volumes to assess the order. This is exactly the check needed to promote the claim from a lattice-spacing-dependent observation to a property of QCD. The reader's CONDITIONAL verdict is therefore appropriate.","tokens_in":96,"tokens_out":9222,"duration_ms":92923,"concrete_test":"Compute the same observables at a finer lattice spacing, e.g. β=4.17, with Nt chosen to keep T≈121 MeV (Nt≈16 given a(β=4.17)<a(β=4.0)) and at the same physical total quark mass around (m_f+m_res)^MS≈3.6 MeV. Run at two spatial volumes (e.g. 32^3 and 48^3) and compare the disconnected chiral susceptibility peak height and Binder cumulant at the pseudocritical point. If at the finer spacing the Binder cumulant remains close to 3 and the susceptibility peak height does not grow with volume, the single-spacing concern is resolved. If the peak grows or B4 moves toward the Z(2)/first-order values, the central claim fails.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central claim—that the Nf=3 transition is a crossover at m_f~4 MeV—is made from data at a single lattice spacing a=0.1361(20) fm (Nt=12) with no continuum extrapolation. The authors themselves note (Sec. 1) that the size of the first-order region depends on the lattice action and spacing, citing Refs. [15–20]. This is not a purely formal issue: at this coarse spacing the residual mass at Ls=16 is am_res=0.00613, i.e. about 9 MeV in physical units, more than twice the quoted physical quark mass. The Ls=32 comparison demonstrates consistency of the disconnected susceptibility at fixed total quark mass, but it does not control O(a^2) discretization effects on the effective potential. Earlier Wilson and staggered studies show that the order can change with spacing, so without a second, finer lattice spacing the observed crossover cannot be taken as evidence about continuum QCD. If the continuum limit turns out to be first-order (or Z(2)) at this mass, the paper's conclusion would be wrong.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper presents an updated lattice study of the Nf=3 QCD phase transition using Möbius domain wall fermions at a single lattice spacing a=0.1361(20) fm (Nt=12, T=121(2) MeV). New data include a 48^3 x 12 x 16 ensemble and 24^3 x 12 x 32 ensembles, together with earlier 24^3 and 36^3 volumes. From the chiral condensate, disconnected chiral susceptibility, Binder cumulant, and histograms of the chiral condensate, the authors conclude that the transition is consistent with a smooth crossover at m_f^MS(2 GeV) ~ 4 MeV. They also propose an explicit subtraction of the additive UV divergence induced by residual chiral symmetry breaking in the domain wall fermion condensate, using parameters C_D and x.","tokens_in":10786,"tokens_out":9681,"duration_ms":94350,"significance":"If taken as a fixed-spacing result, this is a useful data point from a chiral fermion formulation, complementing the staggered and Wilson fermion literature. The addition of a 48^3 volume and the Ls=32 residual-mass control are genuine improvements, and the manuscript is transparent about the modest nature of the evidence. The main significance, however, is limited by the absence of a continuum extrapolation and by the qualitative character of the finite-size analysis; as written, the paper does not settle the continuum nature of the Nf=3 transition, although it provides a relevant observation at this lattice spacing.","major_comments":[{"comment":"The central claim is made from a single lattice spacing, a=0.1361(20) fm (Nt=12), and no continuum extrapolation or second finer lattice spacing is presented. The manuscript itself notes in Sec. 1 that the size of the first-order region depends significantly on the lattice action and spacing (Refs. [15-20]). At this coarse spacing the residual mass at Ls=16 is am_res=0.00613(9), corresponding to about 9 MeV, which is more than twice the quoted physical quark mass. The Ls=32 comparison checks residual-chiral-symmetry effects but does not control O(a^2) discretization effects on the effective potential. I therefore ask that the abstract and summary state explicitly that the crossover observation applies at this lattice spacing and not yet to continuum QCD, or, alternatively, that a finer-spacing test be added.","section":"Sec. 4 and Abstract"},{"comment":"The classification as a crossover rests on Binder cumulant values near 3 and on single-peaked histograms, but no quantitative finite-size scaling analysis is performed. For a weak first-order transition, B4 at finite volume can be close to 3 and the histogram can appear single-peaked if the volume is not large enough to develop phase coexistence. A quantitative comparison with the 3D Z(2) and first-order FSS forms, or at least a fit to the volume dependence of the susceptibility peak height and position, would be needed to distinguish a genuine crossover from a weak first-order transition. As it stands, the conclusion is supported only at the qualitative level.","section":"Sec. 3.4 and Fig. 4"},{"comment":"Equations (4) and (5) are not algebraically equivalent as printed. Equation (4) contains a term C_D m_f without an explicit 1/a^2, while Eq. (5) contains (C_D + C_R a^2)(m_f + m_res)/a^2; these expressions differ unless m_f and m_res are measured in a way that is not stated. Since C_D and x are central to the new UV-subtraction procedure, the lattice or physical units of m_f, m_res, and the condensate must be defined precisely and the equations corrected. Additionally, the extraction of x assumes that the three lowest mass points on 24^3 x 12 x 16 lie in the restored phase; this assumption should be justified with a direct check, for example by showing the disconnected susceptibility at those masses.","section":"Sec. 3.2, Eqs. (4)-(5)"}],"minor_comments":[{"comment":"The word 'transtion' should be corrected to 'transition'.","section":"Abstract and Sec. 4"},{"comment":"The text contains typos such as 'descriaption' and 'perfrom'; a careful proofread is needed.","section":"Sec. 3.2"},{"comment":"The statement that the largest-volume susceptibility is not fitted should be explained; it is unclear why the 48^3 data are excluded from the spline fits.","section":"Sec. 3.3"},{"comment":"The Binder cumulant plot in Fig. 4 does not show visible error bars; if the uncertainties are too small to display, the jackknife or bootstrap procedure should still be described.","section":"Sec. 3.4"},{"comment":"The sentence about obtaining m_res at finite temperature using a spatial source-sink separation is hard to parse and should be rewritten.","section":"Sec. 3.1"}],"recommendation":"major_revision","confidential_remarks":"This is a proceedings-style contribution with a modest claim, and the data production is substantial. My main concern is the single-spacing nature of the crossover statement, which can be addressed by explicit wording or additional analysis. The Eq. (4)-(5) inconsistency should be fixed before publication because it affects the reproducibility of the UV subtraction method."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Yu Zhang and colleagues report a new 48^3x12x16 ensemble at two mass points, an Ls=32 finite-T check, and the first explicit subtraction of the C_D m_f + x m_res / a^2 ultraviolet divergence in the domain-wall chiral condensate. That subtraction is the most useful piece: they show the Ls=16 and Ls=32 condensates, which differ before subtraction, become consistent afterward, and the disconnected susceptibility already matched at equal total quark mass. That is a nontrivial consistency check and a genuine methodological step forward.\n\nThe physics conclusion is modest but honestly presented: at a=0.1361(20) fm, T=121(2) MeV, the Nf=3 transition looks like a crossover at m_f^MS ~ 4 MeV, with Gaussian histograms, Binder cumulant near 3, and only mild volume growth of the susceptibility peak. They explicitly cite the lattice-action and spacing dependence of the first-order region, so they are not overclaiming a continuum statement.\n\nThe soft spots are real but addressable. The whole argument sits at one lattice spacing; the residual mass at Ls=16 is about 9 MeV, more than twice the quoted physical quark mass, and even at Ls=32 it is ~4.7 MeV. The Ls=32 comparison controls residual chiral symmetry breaking but does not control O(a^2) effects on the effective potential. The Binder cumulant is plotted without statistical errors, so the 'near 3' claim cannot be weighed properly. There is no quantitative finite-size scaling fit, and the susceptibility peak heights are compared only visually. The C_D and x subtraction depends on fits and on the assumption that the lightest masses lie in the restored phase; that is plausible given the quoted T_c from Nt=8, but it is an assumption.\n\nThe stress-test worry that the crossover might be a lattice artifact is not contradicted by the paper, but it is also not something the paper claims to resolve: they say 'consistent with crossover' at this spacing and temperature. The weakness is that the abstract and title lean on the QCD phase transition without the qualification. A revised version should state explicitly that no continuum extrapolation is attempted and that the result is one spacing.\n\nWho is this for? Lattice practitioners working on the three-flavor transition or on domain-wall fermions. They get a useful data point and a workable UV subtraction. It deserves a serious referee, but as a proceedings it is probably fine as is; if submitted to a journal, the referee should ask for error bars on B4, a clearer statement of the single-spacing limitation, and ideally the data tables.","headline":"A careful, honest proceedings update that adds a large volume and a useful UV subtraction, but the crossover conclusion rests on a single lattice spacing and no continuum extrapolation.","tokens_in":11349,"tokens_out":2733,"would_cite":true,"duration_ms":26122,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":["12.38.Gc","11.30.Rd"],"model":"deepseek-v4-flash","headline":"Three-flavor QCD's chiral transition at 121 MeV is a smooth crossover at quark mass near 4 MeV.","keywords":["QCD phase transition","chiral crossover","three flavors","Mobius domain wall fermions","disconnected chiral susceptibility","Binder cumulant","residual chiral symmetry breaking","Columbia plot"],"falsifier":"Measure the disconnected chiral susceptibility and Binder cumulant on a $64^{3}$×12×16 lattice at the same quark mass near 4 MeV: if the susceptibility peak height grows roughly in proportion to the spatial volume, or the Binder cumulant moves from 3 toward 1.604 or 1, the crossover interpretation would be refuted; alternatively, a finer lattice with Nt=16 at the same physical temperature that shows a first-order signal would also refute it.","tokens_in":10286,"feed_emoji":"⚛️","tokens_out":6360,"duration_ms":54693,"temperature":0.7,"pith_summary":"This paper uses Möbius domain wall fermions on Nt=12 lattices at a single lattice spacing a=0.1361(20) fm to ask whether the Nf=3 chiral phase transition is first order, second order, or a crossover at light quark masses. The authors add a new $48^{3}$×12×16 volume to previous $24^{3}$ and $36^{3}$ ensembles and examine the chiral condensate, the disconnected chiral susceptibility, and the Binder cumulant. They find that the transition is consistent with a smooth crossover, with the inflection point at m_f^MS(2 GeV) ≈ 4 MeV and T = 121(2) MeV. This matters because the order of the Nf=3 transition constrains the Columbia plot and the possible existence of a critical endpoint; the result agrees with recent staggered and Wilson fermion studies and leaves little room for a first-order region at heavier masses.","feed_headline":"Three-flavor QCD transition is a smooth crossover","feed_subtitle":"A new 48^3 volume shows the chiral susceptibility peak stays flat and the Binder cumulant stays near 3 at quark mass about 4 MeV.","key_machinery":"The argument is carried by three observables computed on Möbius domain wall fermion ensembles: the disconnected chiral susceptibility, whose peak marks the transition and whose volume scaling distinguishes a crossover from a true phase transition; the Binder cumulant of the chiral condensate, with B4 = 3 for crossover, 1 for first order, and 1.604 for the 3d Z(2) universality class; and the distribution of the chiral condensate at the transition point. The chiral condensate itself requires an additive ultraviolet subtraction of the form C_D(m_f + x m_res)/$a^{2}$, with x ≈ −0.6(1) fixed by assuming the condensate vanishes in the chiral limit on the low-mass side. Residual chiral symmetry breaking is quantified by the residual mass m_res, which follows the expected 1/Ls behavior at this strong coupling.","core_discovery":"The central claim is that, at this lattice spacing and temperature, the three-flavor QCD transition is analytic: no first-order signal and no Z(2) critical scaling appears at quark masses down to about 4 MeV. The evidence is that the disconnected chiral susceptibility develops a peak whose height grows only mildly with volume, far less than the linear growth expected for a first-order transition, and the Binder cumulant of the chiral condensate sits near 3, the crossover value, rather than 1 or 1.604. Histograms of the chiral condensate at the transition mass are single-peaked and Gaussian-like at all three volumes. The paper also performs the first explicit subtraction of the ultraviolet-divergent term C_D(m_f + x m_res)/$a^{2}$ from the domain wall chiral condensate, after which Ls=16 and Ls=32 results agree at the same total quark mass; this supports the reliability of the crossover conclusion and clarifies how residual chiral symmetry breaking enters chiral observables.","pith_inferences":["If the crossover extends to the chiral limit, the Nf=3 transition would be a second-order transition of a different universality class or no transition at all, and the Pisarski-Wilczek first-order scenario would be excluded; the present data do not reach the chiral limit, so this is an extrapolation the authors do not make.","The same analysis could be repeated at a finer lattice spacing, such as larger Nt, to test whether the crossover persists; the observed 1/Ls behavior of m_res suggests that Ls must grow as the lattice is refined.","The volume dependence of the susceptibility peak height, if fitted to a scaling form, could be used to place an upper bound on the first-order transition strength at this quark mass.","The procedure for subtracting the ultraviolet divergence in the chiral condensate could be applied to Nf=2+1 simulations, where a physical strange quark mass is present, to separate thermal chiral effects from residual symmetry breaking."],"forward_implications":["If the crossover conclusion holds, the first-order region in the Columbia plot, if it exists at all, must lie at quark masses below roughly 4 MeV.","The result is consistent with recent HISQ and improved Wilson fermion studies that found no first-order transition for pion masses above about 50 to 110 MeV, supporting a small or absent first-order region.","The successful subtraction of the C_D(m_f + x m_res)/a^2 divergence means domain wall fermion chiral condensates can be compared across Ls values, strengthening future studies of chiral observables.","The mild volume dependence of the susceptibility peak near m_f ≈ 4 MeV provides a template for distinguishing crossover from weak first-order behavior on finite lattices.","At this lattice spacing the residual mass obeys a 1/Ls dependence, so Ls must be increased or extrapolated before continuum extrapolations with these fermions become reliable."],"supporting_citations":[{"why":"Supplies the 24^3×12×16 ensembles and the earlier analysis of the chiral transition with these fermions.","marker":"[1]"},{"why":"Supplies the 36^3×12×16 ensembles and the earlier identification of a significant finite-volume effect near the transition.","marker":"[2]"},{"why":"Gives the perturbative prediction of a first-order transition in the Nf=3 chiral limit that this study tests.","marker":"[3]"},{"why":"HISQ study finding no first-order transition for pion masses above 50 MeV, used as a consistency comparison.","marker":"[18]"},{"why":"Improved Wilson fermion study finding no first-order transition for pion masses above 110 MeV, another consistency comparison.","marker":"[20]"},{"why":"Provides the Nt=8 chiral transition temperature Tc=98+3-6 MeV used to assume the chiral condensate vanishes in the restored phase when fixing the ultraviolet subtraction coefficient x.","marker":"[22]"},{"why":"Supplies the continuum-extrapolated scale setting used to convert beta=4.0 to a=0.1361(20) fm and to set the physical temperature.","marker":"[25]"},{"why":"Derives the additive power-divergent contribution x m_res/a^2 in the domain wall fermion chiral condensate, the form used for the subtraction.","marker":"[29]"},{"why":"Provides the 3d Ising Z(2) Binder cumulant value 1.604 that separates crossover from second-order behavior.","marker":"[31]"}],"fun_headline_variants":["Large 48^3 lattice seals 3-flavor QCD crossover","Chiral susceptibility flat despite 48^3 volume","No first-order jump: 3-flavor QCD is crossover","Binder cumulant near 3 confirms smooth transition","Three-flavor QCD: crossover, not first-order"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"All conclusions rest on a single lattice spacing a = 0.1361(20) fm, so the observed crossover could in principle be a lattice artifact rather than a property of continuum QCD.","fun_headline_variants_meta":{"raw":{"variants":["Large 48^3 lattice seals 3-flavor QCD crossover","Chiral susceptibility flat despite 48^3 volume","No first-order jump: 3-flavor QCD is crossover","Binder cumulant near 3 confirms smooth transition","Three-flavor QCD: crossover, not first-order"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000583,"raw_usage":{"total_tokens":2761,"prompt_tokens":979,"completion_tokens":1782,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":595,"completion_tokens_details":{"reasoning_tokens":1698}},"tokens_in":595,"tokens_out":1782,"duration_ms":12094,"temperature":1.0,"reasoning_tokens":1698,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-10T14:14:27.228327+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Measure the disconnected chiral susceptibility and Binder cumulant on a $64^{3}$×12×16 lattice at the same quark mass near 4 MeV: if the susceptibility peak height grows roughly in proportion to the spatial volume, or the Binder cumulant moves from 3 toward 1.604 or 1, the crossover interpretation would be refuted; alternatively, a finer lattice with Nt=16 at the same physical temperature that shows a first-order signal would also refute it.","supporting_citations":[],"review_version":1}