{"id":"1329d006-a1e5-44e5-b0bc-b1d15ee5a792","arxiv_id":"1908.06212","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":7,"one_line_summary":"A hybrid IP-Glasma and hydrodynamics calculation finds that final-state interactions are required for the RHIC small-system flow data, while the initial CGC momentum anisotropy contributes visibly at low multiplicity.","lead":"This paper simulates the smallest collision systems at RHIC using a Color Glass Condensate initial state joined to hydrodynamic evolution, and compares the computed flow to PHENIX data. It concludes that both the initial CGC momentum anisotropy and final-state interactions leave measurable imprints, and it proposes a v2 measurement in d+Au versus Au+Au collisions as a decisive test.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Lost quantum-interference CGC correlations are never included as a no-hydro baseline, so the 'only final-state interactions can reproduce the data' claim is stronger than the evidence; the STAR same-multiplicity system ordering is also opposite to the model.","rationale":"The paper is a serious hybrid calculation with parameters fixed by Au+Au collisions and a broad comparison to PHENIX and STAR data. The strongest part is the demonstration that the classical initial-state epsilon_p has the opposite multiplicity trend from the observed v2, which makes a purely classical initial-state explanation harder. However, the central sentence 'can only be reproduced when final state interactions are present' requires ruling out an initial-state-only calculation that includes the quantum interference terms the paper explicitly says are lost. Since that calculation is not performed and no full no-hydro baseline is provided, the claim is conditional rather than established. The same-multiplicity d+Au versus Au+Au ordering, which the paper proposes as a signature, is also currently opposite to STAR data; the non-flow argument is plausible but unquantified. These are not fatal errors, but they should be explicit conditions on acceptance. My read therefore does not change the reader's CONDITIONAL verdict.","tokens_in":13221,"tokens_out":10820,"duration_ms":118854,"concrete_test":"Recompute v2{2}(dNch/deta) for p+Au, d+Au, 3He+Au, and peripheral Au+Au using the same IP-Glasma configurations, but evaluate the two-particle correlation directly from the color fields via the Glasma-graph/CGC formalism, including the quantum interference terms omitted from T^{\\mu\\nu}, and compare the magnitude and multiplicity trend to the hybrid result in Fig. 6. If the direct-CGC v2 is negligible or has the opposite, decreasing trend with multiplicity, the hydro-necessity claim is supported; if it is comparable in size at low multiplicity, the attribution of the observed v2 to final-state hydrodynamics needs revision. In parallel, reanalyze the STAR data with the same centrality and multiplicity definition and with a subevent cumulant that removes non-flow; if the d+Au > Au+Au ordering persists under non-flow control, the paper's proposed signature is already falsified.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The load-bearing assumption is that the IP-Glasma energy-momentum tensor in Eq. (1), matched to hydrodynamics at tau_init = 0.4 fm/c, is an adequate carrier of CGC initial-state momentum anisotropy, with epsilon_p defined in Eq. (3). The paper itself flags the weak point near Eq. (3): 'the direct CGC calculation of two-particle correlations also includes contributions to the anisotropy from quantum interference effects, which are lost when taking T^{\\mu\\nu} and inserting it into hydrodynamics.' This matters because the evidence for the central claim is the opposite multiplicity trends of epsilon_p and v_2, together with the epsilon_p-v_2 correlations in Fig. 6. A direct CGC two-particle calculation containing the omitted interference terms is the natural no-hydro baseline, but it is never computed here. Without that baseline, 'can only be reproduced when final state interactions are present' is not established against the strongest initial-state-only alternative; the paper only shows that the classical anisotropic stress alone has the wrong multiplicity trend. A second unquantified tension is that at fixed multiplicity the model's system ordering (d+Au > Au+Au) is opposite to the STAR data shown in Fig. 6(a), and the paper invokes non-flow without computing it. Both gaps should be settled before the initial-state imprint is claimed as quantitative.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"This paper presents a hybrid IP-Glasma + MUSIC hydrodynamics + UrQMD transport calculation of multiparticle correlation observables (v2, v3, c2{4}, mean pT, multiplicity distributions) for the RHIC small-system scan: p+p, p+Au, d+Au, and 3He+Au at sqrt(s)=200 GeV. The initial state is the classical Yang-Mills energy-momentum tensor from IP-Glasma, matched to hydrodynamics at tau_init=0.4 fm/c with all parameters previously fixed from Au+Au and HERA constraints. The authors report that the qualitative system/centrality dependence of the measured anisotropies requires final-state hydrodynamic interactions, that the initial CGC momentum anisotropy is correlated with the final v2 (with stronger correlation at low multiplicity), and that d+Au v2 exceeds Au+Au v2 at equal multiplicity, proposed as a discriminating signature.","tokens_in":13479,"tokens_out":10373,"duration_ms":90212,"significance":"If the central claim holds, this is a valuable step in the small-system collectivity debate: it is the first estimate, in a framework with no small-system tuning, of how much of the observed anisotropy can originate in CGC initial-state momentum anisotropy when final-state interactions are described by realistic hydrodynamics. The paper is honest about several of its own limitations (loss of quantum-interference contributions, non-conserving sensitivity tests), and the proposed d+Au vs Au+Au equal-multiplicity measurement is a concrete, falsifiable prediction. The use of publicly available codes (MUSIC, iSS) and the fact that small-system results are genuine predictions rather than fits are additional strengths. However, the 'can only be reproduced when final state interactions are present' statement is stronger than the evidence actually presented, and the same-multiplicity ordering issue with STAR data needs to be resolved before the signature claim is quantitative.","major_comments":[{"comment":"The central claim that the qualitative features of the data 'can only be reproduced when final state interactions are present' is not fully supported, because the only initial-state-only benchmark considered is the classical anisotropic stress epsilon_p of Eq. (3). The paper itself states near Eq. (3) that the direct CGC two-particle correlation 'also includes contributions to the anisotropy from quantum interference effects, which are lost when taking T^{\\mu\\nu} and inserting it into hydrodynamics.' Since a no-hydro baseline containing those quantum-interference terms is never computed, the opposite multiplicity trends of epsilon_p and v2 do not rule out an initial-state-only explanation of the observed increasing v2 with multiplicity. A direct CGC two-particle calculation at the same kinematics (e.g., along the lines of refs. [11,12,16-20]) is the appropriate baseline; without it, the 'only' claim should be weakened, or the baseline must be supplied.","section":"Role of geometry and initial momentum anisotropy (Eq. (3), Fig. 6)"},{"comment":"The model predicts v2{2}(d+Au) > v2{2}(Au+Au) at equal multiplicity, whereas the STAR data shown in Fig. 6(a) display the opposite ordering. The paper attributes this to non-flow in the STAR data but provides no quantitative estimate of that non-flow. Since this ordering is presented as the key testable signature ('a means to reveal effects of the initial state momentum anisotropy'), the tension with existing STAR data must be addressed: either compute non-flow in the model (e.g., through UrQMD or a template-fitted v2{2}) or specify kinematic/rapidity-gap conditions under which the prediction is expected to survive. As written, the signature claim is not yet supported against the available data.","section":"Role of geometry and initial momentum anisotropy (Fig. 6(a); Fig. 4)"},{"comment":"The 90% change in p+Au v2{2} when removing initial flow and/or shear stress comes from initialization schemes that, as the paper states, do not conserve energy and momentum at the switching surface. The magnitude of this change is therefore not a clean measure of the physical importance of the initial flow and viscous stress; it conflates genuine physics with the inconsistency of the matching. The abstract's claim that 'neglecting the initial transverse flow profile or the initial shear stress tensor ... has dramatic effects' should be either backed by a consistent, energy-momentum-conserving projection (e.g., rescaling or re-thermalizing the truncated tensor) or explicitly labeled as a sensitivity test of the implementation rather than a physical estimate. This also affects the interpretation of the 35% change in Au+Au.","section":"Effects of initial flow and viscous stress (Fig. 7)"},{"comment":"The model produces a negative c2{4} in p+Au collisions at multiplicities where the PHENIX data are positive (Fig. 5). Since c2{4} is specifically designed to suppress non-flow, this is a qualitative disagreement for one of the systems in the small-system scan. Combined with the overestimate of v2(pT) and v3(pT) in Fig. 3, the abstract's claim of reproducing 'qualitative features' is too broad; the claim should be qualified to specify which observables and systems are well described. The conclusions already acknowledge the overestimate, but the wording in the abstract and conclusions should be tightened accordingly.","section":"Azimuthal anisotropies (Fig. 5)"}],"minor_comments":[{"comment":"The text cites STAR data as [76,77] while the Fig. 6(a) caption cites [75]; please unify the references for the v2{2} vs multiplicity data.","section":"Fig. 6(a) caption and text"},{"comment":"The conversion N_FVTX_tracks = 1.96 dNch/deta is used to compare with PHENIX c2{4} data; please provide a justification or a reference for this factor, since the comparison in Fig. 5 depends on it.","section":"Footnote 2 (N_FVTX conversion)"},{"comment":"The statement that 'all parameters of the calculation were previously constrained using experimental data on Au+Au collisions' is not literally correct: the IPSat color-charge-density parameters are constrained by HERA data (ref. [57]) and the nucleon-substructure parameters by HERA-informed studies (refs. [71,72]). Please qualify the statement to distinguish parameters constrained by Au+Au data from those inherited from the IP-Glasma/HERA setup.","section":"Abstract"},{"comment":"The y-axis label '105C2{4}' should read '10^5 c2{4}' for clarity.","section":"Fig. 5"},{"comment":"In Eq. (1), the metric signature and the meaning of g^{\\mu\\nu} are not defined; also the text should state explicitly that u^\\mu is timelike and normalized, for readers outside the heavy-ion hydrodynamics subfield.","section":"Framework (Eq. (1))"},{"comment":"The text says 'for 3He we use the same configurations as in [54]'; please clarify whether these configurations include the nucleon hot-spot substructure or only the nucleon positions from the Green's function Monte Carlo wave function.","section":"Role of geometry and initial momentum anisotropy"}],"recommendation":"major_revision","confidential_remarks":"This is a well-executed and timely paper, and the framework is a valuable contribution to the small-system collectivity debate. The major concerns above are fixable within the manuscript's scope: the 'only final-state interactions' claim needs either a no-hydro baseline that includes the quantum-interference terms or an explicit weakening; the STAR same-multiplicity ordering tension needs a non-flow estimate or a kinematic caveat; and the non-conserving sensitivity test needs to be reframed or substantiated with a conserving projection. I do not see citation or novelty problems; the authors appropriately situate their work relative to both CGC-only and hydro-only approaches. The use of public codes and the parameter-fixed nature of the predictions are strengths that should be preserved."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Colleague—read this one; it's the strongest statement yet from the IP-Glasma+hydro camp on the RHIC small-system scan. The new work is the first application of the full T^{\\mu\\nu} initial condition (not just energy density) to p+p, p+Au, d+Au, 3He+Au at 200 GeV, plus a multiplicity-dependent correlation between the CGC momentum anisotropy epsilon_p and final v2, and a clean testable prediction: d+Au v2 > Au+Au v2 at equal multiplicity.\n\nThe execution is careful. All parameters come from the earlier Au+Au constrained setup and HERA structure functions; no small-system data are tuned. The authors are unusually honest: they note the loss of quantum interference contributions when T^{\\mu\\nu} is inserted into hydro, note that the STAR same-multiplicity ordering is opposite to their result, and say the sensitivity tests that remove initial flow or viscous stress break energy-momentum conservation. That candor is real.\n\nThe soft spots are in proportion. The headline claim—'can only be reproduced when final state interactions are present'—is stronger than the evidence. What they actually show is that the classical CGC anisotropic stress alone has the wrong multiplicity trend. That is evidence for final-state effects, but not a proof against the strongest initial-state alternative, namely a direct CGC two-particle calculation that keeps the quantum interference terms they explicitly say are lost. That baseline is never computed. Second, the same-multiplicity system ordering in Fig. 6(a) is opposite to STAR in d+Au vs Au+Au; the non-flow explanation is plausible but unquantified. Third, the model overestimates PHENIX v2 and v3 at pT above 1 GeV and gets the sign of c2{4} wrong in p+Au. The 90% v2 change in p+Au comes from unphysical initial conditions and should be treated as illustrative, not quantitative.\n\nNone of that sinks the paper. The central qualitative point—that final-state interactions are needed to reverse the multiplicity trend—holds up. The prediction for d+Au vs Au+Au at equal multiplicity is a genuinely useful experimental discriminator.\n\nThe absence of code or parameter files is a real reproducibility limitation for a computational paper of this type, but it's common in the field.\n\nThis paper deserves a serious referee. My recommendation: accept after major revision, with the authors required to (1) soften the 'only' claim or supply a genuine no-hydro CGC baseline, (2) quantify the non-flow estimate for the STAR comparison, and (3) discuss the c2{4} sign. I'd bring it to reading group.","headline":"A serious, well-executed hybrid CGC+hydro study of RHIC small systems with genuine predictions, but the central 'only final-state interactions' claim outruns the evidence without a full no-hydro baseline.","tokens_in":14086,"tokens_out":2336,"would_cite":true,"duration_ms":22135,"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":"The flow seen in small collision systems at RHIC can only be reproduced with final-state hydrodynamic interactions, and the CGC initial-state momentum anisotropy leaves an imprint on v2 that grows at low multiplicity.","keywords":["small system scan","Color Glass Condensate","initial momentum anisotropy","elliptic flow","viscous hydrodynamics","RHIC","multi-particle correlations","IP-Glasma"],"falsifier":"Measure $v_2\\{2\\}$ in d+Au and Au+Au at $\\sqrt{s} = 200$ GeV at the same charged-hadron multiplicity, using the same forward-rapidity event plane and mid-rapidity $p_T$ cuts. If d+Au is not systematically above Au+Au at equal multiplicity, the proposed signature of initial momentum anisotropy is refuted. A cheaper test is to rerun this hybrid with initial flow and shear removed; the paper predicts $v_2$ changes by up to 90% in p+Au, so a much smaller change would contradict the model.","tokens_in":1860,"feed_emoji":"⚛️","tokens_out":2184,"duration_ms":83009,"temperature":0.7,"pith_summary":"The paper asks whether the azimuthal momentum anisotropies seen in small collision systems at RHIC (p+p, p+Au, d+Au, 3He+Au) come from the initial state within the Color Glass Condensate effective theory or from final-state interactions. It builds a hybrid calculation that includes both: the CGC supplies a full energy-momentum tensor event by event, viscous hydrodynamics evolves it, and hadronic transport handles the dilute final stage, with all parameters previously fixed by Au+Au data. The central conclusion is that the qualitative features of the data, such as the system and centrality dependence of the charged-hadron momentum anisotropy, can only be reproduced when final-state interactions are present. Quantitative agreement, however, also requires the full initial-state momentum anisotropy, and the paper shows that this initial anisotropy correlates with the observed elliptic flow in all small systems, most strongly at low multiplicity. The paper identifies a same-multiplicity comparison of v2 in d+Au and Au+Au collisions at RHIC as the way to expose this initial-state effect.","feed_headline":"Small-system flow needs final-state physics, not just geometry","feed_subtitle":"The CGC initial-state imprint on v2 grows at low multiplicity; equal-multiplicity d+Au vs Au+Au could expose it.","key_machinery":"The load-bearing object is the full classical Yang-Mills energy-momentum tensor $T^{\\mu\\nu}_{\\mathrm{CYM}}$ computed event by event at a proper time $\\tau_{\\mathrm{init}} = 0.4$ fm/c and fed directly into the hydrodynamic initial conditions. This tensor carries the initial-state momentum anisotropy, and the paper isolates its role by decomposing it into energy density, flow velocity $u^\\mu$, shear stress $\\pi^{\\mu\\nu}$, and an effective bulk pressure $\\Pi = \\varepsilon/3 - P_{\\mathrm{lat}}$. Removing any one of these pieces changes the final $v_2$, with the largest effect in p+Au collisions, where $v_2$ changes by up to 90% when initial flow and shear are dropped.","core_discovery":"The paper claims that both initial-state CGC momentum anisotropy and final-state hydrodynamic response are needed to describe the RHIC small-system scan, but that final-state interactions are indispensable: no purely initial-state picture reproduces the observed v2 trends. Within the hybrid calculation, the initial momentum anisotropy epsilon_p, defined from the classical Yang-Mills energy-momentum tensor, is anticorrelated with multiplicity while the final v2 rises with multiplicity, showing that hydrodynamics reverses the initial-state trend. The paper also finds that the magnitude and orientation of epsilon_p are correlated with the final v2 in all small systems, with the correlation increasing toward low multiplicity and essentially vanishing in central Au+Au collisions. At equal multiplicity, the calculation predicts d+Au v2 to exceed Au+Au v2, an effect the authors attribute to the initial momentum anisotropy rather than to geometry or mean transverse momentum differences.","pith_inferences":["The paper itself notes that direct CGC two-particle correlation calculations include quantum interference contributions that are lost when only the energy-momentum tensor is inserted into hydrodynamics; if those contributions are significant at low multiplicity, the quantitative correlation values could shift even if the qualitative picture holds.","The proposed same-multiplicity d+Au versus Au+Au measurement would also discriminate between CGC-style initial momentum correlations and purely geometric hydrodynamic explanations, since the two systems have similar eccentricities but ordered opposite to the predicted v2 difference.","At LHC energies the paper argues hydrodynamics gains relative importance because fireballs live longer, so an extension to p+Pb at 5.02 TeV would likely predict a smaller same-multiplicity initial-state imprint than at RHIC.","The strong sensitivity of v2 to initial flow and shear in p+Au suggests that simpler energy-density-only initial conditions used in many event generators understate the role of early-time dynamics in small collision systems."],"forward_implications":["Final-state hydrodynamic response is necessary to reproduce the qualitative system and centrality dependence of the measured momentum anisotropies in small systems.","Any initial-state model for small collision systems must include the CGC momentum anisotropy, not just the spatial energy-density distribution, if quantitative v2 is the goal.","At multiplicities below roughly ten charged hadrons per unit rapidity, initial momentum anisotropy magnitude and direction are correlated with final elliptic flow; in Au+Au collisions above about 50% centrality the correlation disappears.","A same-multiplicity comparison of v2 in d+Au and Au+Au at RHIC should show d+Au above Au+Au if this framework is correct, providing a testable signature of initial-state momentum anisotropy.","The v2(pT) ordering between p+Au, d+Au, and Au+Au at matched multiplicity follows the expected geometric ordering, but the integrated v2 difference between d+Au and Au+Au is attributed to the initial momentum anisotropy."],"supporting_citations":[{"why":"Supplies the hydrodynamic setup and the Au+Au-constrained parameter set that fixes all free parameters of this calculation.","marker":"[47]"},{"why":"Establishes the hybrid framework that couples the full IP-Glasma energy-momentum tensor to hydrodynamics, which this paper extends to the RHIC small-system scan.","marker":"[37]"},{"why":"Provides the PHENIX small-system scan data on v2 and v3 that the calculation is compared against.","marker":"[44]"},{"why":"Defines the IP-Glasma initial-state model that computes the classical Yang-Mills fields and the energy-momentum tensor used as hydrodynamic input.","marker":"[38,39]"},{"why":"Prior IP-Glasma plus Boltzmann study showing that initial-state correlations survive at high pT and at low multiplicity, which motivates the present hybrid study.","marker":"[40]"},{"why":"Independent CGC calculation giving the same ordering of initial momentum anisotropy between p+Au and d+Au, supporting the interpretation of the final-state results.","marker":"[18]"},{"why":"Describes the UrQMD hadronic transport used for the low-density final stage after hydrodynamic evolution.","marker":"[45,46]"},{"why":"Supports the definition of the initial momentum anisotropy epsilon_p as a proxy for the purely initial-state v2.","marker":"[74]"}],"fun_headline_variants":["Hydro reverses CGC anisotropy to produce observed v2","Sole initial state fails; final state hydro is essential","CGC + hydro both needed; final state is the decider","d+Au vs Au+Au at same multiplicity could expose CGC","CGC anisotropy matters most at low multiplicity"],"cache_read_input_tokens":16128,"weakest_assumption_plain":"The calculation assumes that the classical energy-momentum tensor from the CGC, matched to hydrodynamics at $\\tau_{\\mathrm{init}} = 0.4$ fm/c, carries all the initial-state momentum anisotropy that matters for the final $v_2$, while the quantum interference pieces of genuine two-particle CGC correlations are set aside.","fun_headline_variants_meta":{"raw":{"variants":["Hydro reverses CGC anisotropy to produce observed v2","Sole initial state fails; final state hydro is essential","CGC + hydro both needed; final state is the decider","d+Au vs Au+Au at same multiplicity could expose CGC","CGC anisotropy matters most at low multiplicity"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000671,"raw_usage":{"total_tokens":3077,"prompt_tokens":982,"completion_tokens":2095,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":598,"completion_tokens_details":{"reasoning_tokens":2013}},"tokens_in":598,"tokens_out":2095,"duration_ms":13133,"temperature":1.0,"reasoning_tokens":2013,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-14T12:53:00.442160+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Measure $v_2\\{2\\}$ in d+Au and Au+Au at $\\sqrt{s} = 200$ GeV at the same charged-hadron multiplicity, using the same forward-rapidity event plane and mid-rapidity $p_T$ cuts. If d+Au is not systematically above Au+Au at equal multiplicity, the proposed signature of initial momentum anisotropy is refuted. A cheaper test is to rerun this hybrid with initial flow and shear removed; the paper predicts $v_2$ changes by up to 90% in p+Au, so a much smaller change would contradict the model.","supporting_citations":[],"review_version":1}