{"id":"9ae36e45-5f48-45c3-844a-b9e62eab41e7","arxiv_id":"1908.09552","paper_version":2,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":2,"one_line_summary":"Continuum-extrapolated lattice QCD screening masses show vector-axial-vector degeneracy at Tpc and scalar-pseudoscalar degeneracy near 1.3Tpc, while J=0 and J=1 channels remain split up to 2.5 GeV.","lead":"By simulating quarks and gluons on a lattice, the authors measured how the screening masses of mesons change as the temperature of the quark-gluon plasma rises. The results map which symmetries of the strong force return at high temperature and test the perturbative description of hot QCD.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The 1.3 Tpc UA(1) conclusion rests on the pathological staggered scalar channel; the phase-space and susceptibility-continuum arguments are asserted, not demonstrated.","rationale":"The reader's weakest assumption is the right one. The paper's central claim has two parts: chiral SU(2) restoration at Tpc, supported by V/AV degeneracy, and UA(1) restoration at ≈1.3Tpc, supported by S/PS degeneracy and by m_s^2(χπ−χa0). The first part is backed by a controlled five-Nτ continuum extrapolation and is not threatened by the staggered scalar pathology. The second part is exactly where the paper admits its operator is problematic (Sec. IV C). The phrase 'might be closed around Tpc due to lack of phase space' is a plausibility argument, not a demonstration; and the susceptibility argument, while using a standard renormalized quantity, is presented for one flavor channel and is not checked against the two-pion contamination that the same section identifies. This is an internally acknowledged gap in the evidence chain, not a disagreement with outside consensus; indeed the result agrees with chiral-fermion calculations, which is why I would not reject the paper. The conditional verdict is appropriate: the conclusion is credible but not established by the quantitative check that would be needed. I keep the reader's verdict unchanged.","tokens_in":26626,"tokens_out":12959,"duration_ms":145718,"concrete_test":"On the same HISQ ensembles, recompute the ¯ud scalar screening mass and χa0 using a non-local scalar operator that projects onto the physical taste (as used in Sec. IV B and Ref. [32]) instead of the local M1, for T = 0.16–0.25 GeV and Nτ = 8, 10, 12, 16. Determine where the resulting S-PS screening masses become degenerate and where m_s^2(χπ−χa0) crosses zero. If either temperature moves by more than ≈20 MeV from the values quoted in Sec. IV C, the reported 1.3Tpc UA(1) restoration is not cleanly separated from the staggered two-pion artifact.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The load-bearing point is the light-light scalar channel used for the UA(1) claim. The paper admits in Sec. IV C (and Fig. 3) that at finite lattice spacing the local scalar operator M1 yields a T=0 mass of 2mπ instead of the physical a0, and that the artifact would only cancel if the continuum limit were taken before extracting the mass. The S-PS screening-mass degeneracy at T≈200 MeV is nevertheless interpreted as effective UA(1) restoration, with the supporting argument that the unphysical ππ channel is 'possibly closed' near Tpc by phase space. The text then switches to the susceptibility difference m_s^2(χπ−χa0), Eq. (5)/Fig. 8, as a cleaner quantity. That switch is reasonable in principle, but no quantitative check is shown that (i) the two-pion overlap of M1 is negligible in the T≈1.3Tpc window, or (ii) the integrated staggered susceptibility χa0 is free of the same taste-breaking two-pion contamination after summing over z and extrapolating to the continuum. If the apparent S-PS degeneracy and the zero of the susceptibility difference are both consequences of the unphysical ππ state, the paper's 1.3Tpc UA(1) claim loses its primary support. The chiral SU(2) conclusion from V/AV degeneracy at Tpc is independent of this issue. Agreement with overlap/domain-wall calculations in the literature is encouraging, but it does not replace a check of this specific staggered extraction.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper presents a lattice QCD determination of mesonic screening masses in the temperature range 140 MeV to 2.5 GeV using (2+1)-flavor HISQ fermions with physical strange quark mass and light quark masses corresponding to pion masses of 140 and 160 MeV. Screening masses are extracted from spatial correlation functions for scalar, pseudoscalar, vector, and axial-vector channels in the light-light, light-strange, and strange-strange sectors, using multi-state fits with AICc-based model selection, and continuum extrapolations are performed with five lattice spacings (Nτ = 6–16) for T ≲ 1 GeV. The main physical conclusions are: (i) the ud vector and axial-vector screening masses become degenerate at or very close to Tpc, indicating effective chiral SU(2) restoration; (ii) the ud scalar and pseudoscalar screening masses become degenerate only around T ≈ 200 MeV (about 1.3 Tpc), which is interpreted as evidence for effective UA(1) restoration; and (iii) at high temperatures the screening masses overshoot the free-field value 2πT and the deviations are compared qualitatively with EQCD predictions.","tokens_in":27010,"tokens_out":6752,"duration_ms":64276,"significance":"If the conclusions hold, this is a valuable and fairly comprehensive lattice result. The chiral SU(2) part of the claim—vector/axial-vector degeneracy at Tpc—appears robust across lattice spacings and is supported by the continuum extrapolation; it also agrees with the broader literature. The high-temperature comparison with EQCD provides a useful benchmark, and the tabulated continuum-extrapolated screening masses in Appendix C are a useful resource for the community. The paper also contains a detailed description of a difficult multi-state fitting procedure with AICc selection, which is a strength in terms of methodological transparency. The main novelty is the UA(1) restoration temperature inferred from the scalar and pseudoscalar channels, but this is precisely the part of the analysis that is most sensitive to the known staggered-fermion artifact in the ud scalar channel. Because the paper's supporting arguments are qualitative rather than quantitative, the UA(1) conclusion must be viewed as tentative until the artifact question is resolved; the chiral SU(2) and high-temperature parts stand on their own.","major_comments":[{"comment":"The central claim of effective UA(1) restoration at about 1.3 Tpc rests on the degeneracy of the light-light scalar and pseudoscalar screening masses, but the scalar channel is the one contaminated by the unphysical two-pion state arising from staggered taste mixing. As the paper itself states in Sec. IV C and in the caption of Fig. 3, at finite lattice spacing the local M1 operator gives a mass of 2mπ rather than the physical a0, and this artifact would cancel only if the continuum limit were taken before extracting the mass. In the present analysis the mass is extracted at finite lattice spacing and then extrapolated linearly in 1/Nτ^2, so the artifact does not cancel by construction. The argument that the unphysical decay channel is 'possibly closed' around Tpc due to lack of phase space is qualitative, and no quantitative estimate is provided for the overlap of the M1 operator with the two-pion state in the T ≈ 200 MeV window. The paper should either supply such an estimate (for example by comparing with a non-staggered calculation at one or two temperatures, or by studying the correlator with the continuum limit taken before the spectral extraction) or explicitly present the S-PS degeneracy as a suggestive, but not established, signal of UA(1) restoration.","section":"§IV C (Fig. 7)"},{"comment":"The susceptibility difference m_s^2 (χ_π − χ_a0) is introduced as a cleaner observable for UA(1) restoration, but the paper does not demonstrate that the integrated staggered scalar susceptibility χ_a0 is free of the same two-pion taste contamination after summation over z and continuum extrapolation. The text states that this quantity has a convergent continuum limit, but no separate continuum limits of χ_π and χ_a0 are shown, and the phase-space argument used for the screening mass is not directly applicable to the integrated quantity. Since the zero crossing of the susceptibility difference at T ≈ 200 MeV is used to support the 1.3 Tpc restoration temperature, the paper should provide a quantitative check that the two-pion contribution is absent or negligible in the continuum limit of χ_a0, or should downgrade this conclusion accordingly.","section":"§IV C, Eq. (5), Fig. 8"}],"minor_comments":[{"comment":"The sentence 'the unphysical contribution cancels out if one would take the continuum limit for the correlator first' appears to describe an analysis that is not performed here; please clarify whether this statement refers to a theoretical property of the discretization or to a check that was actually carried out on the screening masses or susceptibilities.","section":"§IV C"},{"comment":"The channel labels 'A V S P' at the top of each panel are not expanded in the captions; please spell out axial vector, vector, scalar, pseudoscalar.","section":"Fig. 5 and Fig. 7"},{"comment":"The AICc-based fitting and plateau selection are described in the main text, but details of the fit ranges, the number of states used for each channel and temperature, and the stability checks against varying the fit interval are delegated to a PhD thesis [46]. Please provide at least a short summary of the typical fit ranges and systematic errors in the paper, so that the reader can assess the quoted uncertainties.","section":"§III B and Fig. 1"}],"recommendation":"major_revision","confidential_remarks":"The UA(1) claim is the main novel result and is the part most at risk; the outcome of the revision hinges on whether the requested quantitative check can be provided. The chiral SU(2) result and the high-temperature EQCD comparison are solid and would support publication even if the UA(1) claim is softened to a suggestive rather than definitive conclusion."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Short version: this is a careful lattice QCD paper that delivers something genuinely new—continuum-extrapolated meson screening masses for light, strange, and mixed flavor channels from 140 to 1000 MeV using HISQ with physical quark masses and five lattice spacings, plus single-spacing results up to 2.5 GeV. The V/AV degeneracy in the ud sector at Tpc is robust and is the paper's cleanest physics result. The UA(1) claim from S/PS degeneracy at about 1.3 Tpc is plausible but softer than the abstract suggests.\n\nWhat is good: the data tables in Appendix C will be a reference. The scale-setting update using fK is careful, and the fit methodology with AICc selection and bootstrap errors is reasonable. The paper is also honest: it states explicitly that the local staggered scalar operator reproduces 2mπ rather than the a0 at T=0 (Fig. 3), and it flags the unphysical ππ coupling in Sec. IV C. That honesty earns credit.\n\nWhere I agree with the conditional verdict: the stress-test note is on target. The S/PS degeneracy around 200 MeV is read as effective UA(1) restoration, but the scalar channel at finite lattice spacing is dominated by the unphysical two-pion state. The paper's phase-space argument that the channel 'might be closed' is qualitative. It then moves to the susceptibility difference m_s^2(χπ−χa0), which is a reasonable quantity in principle, but there is no quantitative check that the two-pion overlap is negligible in that window or that the integrated χ_a0 is free of the same contamination after summing over z. If the apparent degeneracy is a consequence of that staggered artifact, the 1.3 Tpc claim loses its primary support. Agreement with chiral-fermion and eigenvalue-based calculations in the literature is encouraging, but it does not replace a check of this specific staggered extraction.\n\nMinor soft spots: the high-temperature comparison to EQCD is only qualitative, and above 1 GeV there is no continuum extrapolation because only Nτ=8 data exist. The manual plateau selection, despite AICc, leaves some fit-selection systematic unquantified. These are not deal-breakers.\n\nBottom line: send it to peer review. The right outcome is probably publication after the UA(1) section is either strengthened with a quantitative check or explicitly reframed as suggestive. I would cite the screening mass tables and would bring it to a reading group.","headline":"Solid new reference data for screening masses; chiral SU(2) restoration near Tpc is robust, but the UA(1) claim at 1.3 Tpc rests on a staggered scalar channel the paper itself flags as pathological, so treat that conclusion as conditional.","tokens_in":27519,"tokens_out":3412,"would_cite":true,"duration_ms":35675,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":["11.30.Rd","12.38.Gc","12.38.Mh"],"model":"deepseek-v4-flash","headline":"In (2+1)-flavor QCD, light vector and axial-vector screening masses become degenerate at the chiral crossover temperature, while scalar and pseudoscalar screening masses become degenerate only at about 1.3 times that temperature.","keywords":["lattice QCD","screening masses","chiral symmetry restoration","axial U(1) anomaly","HISQ action","continuum extrapolation","quark-gluon plasma"],"falsifier":"Compute the scalar and pseudoscalar screening masses at $T\\simeq 200$ MeV with a fermion discretization that has no staggered two-pion artifact, such as chiral fermions, and check whether they are degenerate; a split there would push the effective $U_A(1)$ restoration temperature higher than $1.3\\,T_{pc}$, while degeneracy already at $T_{pc}$ would disprove the paper's hierarchy.","tokens_in":26429,"feed_emoji":"⚛️","tokens_out":13835,"duration_ms":121566,"temperature":0.7,"pith_summary":"Using (2+1)-flavor lattice QCD with physical strange and near-physical light quark masses, this paper computes meson screening masses from 140 MeV to 2.5 GeV and extracts continuum limits up to 1 GeV. The central result is a symmetry-restoration hierarchy in the light-light channel: vector and axial-vector screening masses become degenerate at or very close to the chiral crossover temperature $T_{pc} \\simeq 156.5$ MeV, while scalar and pseudoscalar screening masses stay split until about $1.3\\,T_{pc}$ ($T \\sim 200$ MeV). The first degeneracy is the signal of effective $SU_L(2)\\times SU_R(2)$ restoration; the second is the signal of effective $U_A(1)$ restoration. Because the two temperatures differ, the paper concludes that the anomalous axial symmetry is still broken at the chiral crossover and only effectively restored well above it. At high temperature, screening masses sit above the free-quark value $2\\pi T$, with $J=0$ and $J=1$ channels remaining non-degenerate up to the highest temperature studied.","feed_headline":"Vector and axial mesons merge at Tpc; scalars lag to 1.3 Tpc","feed_subtitle":"Lattice QCD with continuum extrapolation shows chiral symmetry is restored before the axial U(1) symmetry.","key_machinery":"The machinery is the screening correlator $G_\\Gamma(z,T)$, whose exponential decay defines the screening mass $m_\\Gamma(T)$. Degeneracies among these masses act as symmetry order parameters: vector and axial-vector degeneracy signals $SU_L(2)\\times SU_R(2)$ restoration, and scalar and pseudoscalar degeneracy signals $U_A(1)$ restoration. Because the staggered scalar correlator contains an unphysical two-pion state at finite lattice spacing, the paper uses the continuum-extrapolated susceptibility difference $m_s^2(\\chi_\\pi-\\chi_{a_0})$ as the quantitative $U_A(1)$ probe, together with a phase-space argument that the spurious decay channel closes near $T_{pc}$. Continuum limits are obtained from five lattice spacings ($N_\\tau = 6,8,10,12,16$) through a combined spline interpolation in temperature and a linear extrapolation in $1/N_\\tau^2$.","core_discovery":"The discovery is a temperature ordering of symmetry restoration read off from continuum-extrapolated screening masses. For the light-light ($\\bar{u}d$) sector, the vector ($\\rho$) and axial-vector ($a_1$) screening masses become degenerate right at the pseudo-critical temperature $T_{pc}$; this is the signature of effective $SU_L(2)\\times SU_R(2)$ chiral restoration. The scalar and pseudoscalar channels ($a_0$ and $\\pi$), which would be degenerate under the anomalous $U_A(1)$ symmetry, become degenerate only near $T\\sim 200$ MeV $\\simeq 1.3\\,T_{pc}$, and the continuum-extrapolated susceptibility difference $m_s^2(\\chi_\\pi-\\chi_{a_0})$ vanishes at about that temperature. The same degeneracy sequence appears in the $\\bar{u}s$ and $\\bar{s}s$ sectors at progressively higher temperatures. The paper also establishes that screening masses are larger than $2\\pi T$ at high temperature and that different angular-momentum channels remain split up to 2.5 GeV, concluding that the free-quark degeneracy is only approached asymptotically.","pith_inferences":["This suggests that the gauge-field fluctuations responsible for $U_A(1)$ breaking survive the chiral crossover and are suppressed only around $1.3\\,T_{pc}$; measuring the topological susceptibility on the same ensembles across temperature would test this directly.","Because screening masses are extracted from spatial correlators, they reflect thermodynamics rather than pole masses; the same hierarchy in temporal correlators or spectral functions would be a stronger statement, and remains an open check.","The mass-ordering pattern suggests a systematic extension: screening masses for charm or bottom channels along the same line of constant physics should show degeneracy temperatures that continue to rise with valence quark mass, sharpening the empirical map of symmetry restoration."],"forward_implications":["Effective $SU_L(2)\\times SU_R(2)$ restoration in the light-light sector is realized at $T_{pc}$, while effective $U_A(1)$ restoration is delayed to $T\\simeq 1.3\\,T_{pc}$.","The continuum-extrapolated, renormalized susceptibility difference $m_s^2(\\chi_\\pi-\\chi_{a_0})$ is non-zero at the crossover and vanishes near $T\\sim 200$ MeV, giving a quantitative signal for the $U_A(1)$ restoration temperature.","In the $\\bar{u}s$ and $\\bar{s}s$ sectors the same degeneracies occur at higher temperatures, so heavier quark masses postpone effective symmetry restoration.","At high temperature, screening masses overshoot $2\\pi T$ and the $J=0$ and $J=1$ channels remain separated up to 2.5 GeV, so the free-theory limit is reached only asymptotically.","The leading-order EQCD calculation gives a spin-independent, positive correction that describes the vector and axial-vector overshoot qualitatively but lies above the scalar and pseudoscalar data, so higher-order spin-dependent corrections are needed."],"supporting_citations":[{"why":"defines the spatial meson correlator, the screening mass, and the free-theory limit $2\\pi T$ used as the high-temperature reference.","marker":"[5]"},{"why":"supplies the pseudo-critical temperature $T_{pc}=156.5(1.5)$ MeV that fixes the crossover scale of the analysis.","marker":"[6]"},{"why":"the earlier staggered-fermion screening-mass study whose $U_A(1)$ degeneracy at $T\\gtrsim1.3T_{pc}$ this paper extends with continuum extrapolations.","marker":"[12]"},{"why":"provides the HISQ scale and taste-splitting analysis whose scale parametrization is updated here.","marker":"[40]"},{"why":"provides the gauge ensembles and the line of constant physics used for the lattice data.","marker":"[41]"},{"why":"describes the automated multi-state fit and AICc plateau selection used to extract the screening masses.","marker":"[46]"},{"why":"explains the staggered two-pion artifact in the scalar channel and its cancellation in the continuum limit.","marker":"[53–55]"},{"why":"gives the EQCD potential needed for the leading-order perturbative screening-mass correction compared at high temperature.","marker":"[72]"}],"fun_headline_variants":["Chiral partners merge at Tpc, axial U(1) at 1.3 Tpc","Lattice QCD: rho-a1 merge at Tpc, a0-pi at 1.3Tpc","Continuum QCD: chiral restored at Tpc, U_A(1) at 1.3Tpc","Rho and a1 degenerate at Tpc; a0 and pi at 1.3Tpc","Symmetry sequence: rho-a1 at Tpc, a0-pi at 1.3Tpc"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The load-bearing assumption is that the unphysical two-pion contribution in the staggered scalar correlator does not survive the continuum limit and is kinematically closed near $T_{pc}$, so that the observed scalar-pseudoscalar splitting and the vanishing of $m_s^2(\\chi_\\pi-\\chi_{a_0})$ at $T\\simeq 200$ MeV are genuine $U_A(1)$ effects rather than lattice artifacts.","fun_headline_variants_meta":{"raw":{"variants":["Chiral partners merge at Tpc, axial U(1) at 1.3 Tpc","Lattice QCD: rho-a1 merge at Tpc, a0-pi at 1.3Tpc","Continuum QCD: chiral restored at Tpc, U_A(1) at 1.3Tpc","Rho and a1 degenerate at Tpc; a0 and pi at 1.3Tpc","Symmetry sequence: rho-a1 at Tpc, a0-pi at 1.3Tpc"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.00184,"raw_usage":{"total_tokens":7223,"prompt_tokens":924,"completion_tokens":6299,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":540,"completion_tokens_details":{"reasoning_tokens":6160}},"tokens_in":540,"tokens_out":6299,"duration_ms":41335,"temperature":1.0,"reasoning_tokens":6160,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-14T11:07:57.045221+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Compute the scalar and pseudoscalar screening masses at $T\\simeq 200$ MeV with a fermion discretization that has no staggered two-pion artifact, such as chiral fermions, and check whether they are degenerate; a split there would push the effective $U_A(1)$ restoration temperature higher than $1.3\\,T_{pc}$, while degeneracy already at $T_{pc}$ would disprove the paper's hierarchy.","supporting_citations":[{"cited_title":"Use the pa- rameters from steps 3 and 4 as initial guess","cited_arxiv_id":null,"evidence_quote":"defines the spatial meson correlator, the screening mass, and the free-theory limit $2\\pi T$ used as the high-temperature reference."},{"cited_title":"Having developed a method to perform automated multiple state fits, we still have to find which set of fit parameters is the most reasonable one for a given fit interval","cited_arxiv_id":null,"evidence_quote":"supplies the pseudo-critical temperature $T_{pc}=156.5(1.5)$ MeV that fixes the crossover scale of the analysis."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"provides the HISQ scale and taste-splitting analysis whose scale parametrization is updated here."}],"review_version":1}