{"id":"903e6ac6-ad86-4145-9352-f3e18f528c94","arxiv_id":"2411.09323","paper_version":1,"verdict":"ACCEPT","confidence":"MODERATE","novelty_score":8.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":3,"one_line_summary":"Baryon and meson elliptic flow coefficients group and split at intermediate transverse momentum in pp and p-Pb collisions, signaling possible partonic flow in small systems.","lead":"ALICE reports the first observation of baryon and meson elliptic flow grouping in high-multiplicity proton-proton and proton-lead collisions at the LHC. The grouping, previously seen only in heavy-ion collisions, is consistent with a flowing partonic system in small collision systems.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Template-fit and event-class matching: the baryon-meson v2 splitting could be partly an artifact of multiplicity-dependent non-flow unless the LM template matching is validated.","rationale":"I agree with the reader that the template-fit assumption is the weakest point, and I would not reject the paper. The measurement is real, the analysis is thorough, and the systematic uncertainties are documented. However, the paper's headline claim is a new observation of baryon-meson grouping/splitting, and its interpretation as partonic flow is only as strong as the non-flow subtraction. The template fit would be falsified if non-flow correlations are multiplicity dependent in a way not captured by a single scale F. Since the paper itself shows (via String-Shoving PYTHIA) that the template method can yield negative v2 for an alternative non-flow model, the method's output is model-shape dependent, and the positive v2 and the baryon-meson splitting in data could in principle be partly an artifact of the template shape. The proposed Nch-matched template test is a specific, feasible, and standard robustness check that would settle this. Thus the verdict should be CONDITIONAL: accept the paper's data and observation of a pattern, but with the explicit condition that the template-matching check confirms the splitting; otherwise the claim of partonic flow is overstated. This is a modest adjustment from ACCEPT, not a rejection.","tokens_in":25957,"tokens_out":3630,"duration_ms":38538,"concrete_test":"One decisive check: re-run the extraction with the LM template built from events whose Nch distribution is matched to the HM event class (e.g., the same Nch bins or an Nch-weighted average of LM and mid-multiplicity events), rather than the fixed low-multiplicity Nch<20 sample, and recompute v2 and the baryon-meson difference at 3<pT<8 GeV/c. If the grouping and ~5-sigma splitting survive with a matched template, the non-flow concern is largely closed; if the splitting is reduced to below ~3-sigma or changes sign, the central claim is not robust.","verdict_should_be":"CONDITIONAL","load_bearing_attack":"The reader's weakest assumption is correct and is the most load-bearing point. The central claim is the first observation of baryon-meson v2 grouping and ~5-sigma splitting in pp and p-Pb at intermediate pT, interpreted as evidence for partonic flow. The extraction relies entirely on Eq. A.1 (template fit): Y_HM = F*Y_LM + G*(1 + sum 2VnDelta cos(n*DeltaPhi)). This assumes the HM correlation function is a scaled LM non-flow template plus a flow component, i.e., that the shape of non-flow correlations is identical in HM and LM and only its amplitude scales by F. The paper's non-flow residual estimate (<1% for pT>0.6 GeV/c) comes from applying the template fit to PYTHIA 8, which contains no collective flow; it cannot validate the template method in real data where a collective component is present. The LM event classes (Nch<20; 60-100% V0A) are not matched to the HM classes in Nch (<Nch>≈35), so if jet fragmentation or resonance-decay correlations per particle grow with multiplicity in a species-dependent way, the single scale F cannot absorb the difference and the extracted v2 becomes biased. The paper's own observation that String-Shoving PYTHIA yields negative v2 after template fit demonstrates that the method is sensitive to the detailed shape and normalization of the LM template, not just to a global scaling, and it is exactly this sensitivity that could generate a spurious ~1-sigma grouping and ~5-sigma splitting: e.g., a larger baryon-jet-correlation component in HM than in LM would leave a species-dependent positive residual. Therefore the strength of the observation, and the partonic-flow interpretation, hinges on demonstrating that the template subtraction is robust to realistic multiplicity-dependent non-flow shapes.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The ALICE Collaboration presents pT-differential elliptic flow coefficients v2 for identified mesons (pi^pm, K^pm, K0S) and baryons (p+pbar, Lambda+Lambdabar) in high-multiplicity pp collisions at sqrt(s)=13 TeV and p-Pb collisions at sqrt(s_NN)=5.02 TeV, using two-particle correlations with a large pseudorapidity separation and a template-fit subtraction of non-flow contributions. The paper reports, for the first time, a baryon-meson grouping (within about 1 sigma) and splitting (about 5 sigma) at intermediate pT (about 3--8 GeV/c) in both small systems, similar to the pattern observed in Pb-Pb collisions. The data are compared with the Hydro-Coal-Frag model, which includes hydrodynamics, quark coalescence, and fragmentation, and with a Hydro-Frag model without coalescence; only the former reproduces the grouping and splitting. On this basis the authors conclude that the created system in high-multiplicity pp and p-Pb collisions includes a stage with collectively flowing partons.","tokens_in":26281,"tokens_out":4182,"duration_ms":47053,"significance":"If the observation is robust, this is the first identified-particle measurement of the baryon-meson v2 grouping and splitting in small collision systems, and it provides qualitatively new evidence in the debate about QGP-like collectivity in pp and p-Pb collisions. The paper has substantial experimental strengths: very large event samples, a Bayesian PID approach with high purity, a three-subevent correlation method with large |Delta-eta| gaps, and a systematic-uncertainty treatment that includes event-selection, track, PID, and template variations. The model comparison is also valuable, and the authors correctly note that the Hydro-Coal-Frag parameters were tuned to pT spectra rather than to v2, making the comparison semi-independent. The central limitation is that the v2 extraction, and hence the claimed grouping and splitting, rests on the template-fit ansatz of Eq. (A.1), whose validity against multiplicity-dependent non-flow is not fully demonstrated; this is the main load-bearing point that needs strengthening before the claim can be accepted at face value.","major_comments":[{"comment":"The template-fit method assumes that the high-multiplicity (HM) correlation function is a scaled low-multiplicity (LM) non-flow template plus a flow modulation, i.e. that all non-flow correlations have the same shape in HM and LM and differ only by a global scale F. The LM and HM event classes used here are not matched in Nch (LM requires Nch<20, HM requires Nch>25 and yields <Nch>~35), and if jet-fragmentation or resonance-decay correlations per particle grow with multiplicity in a species-dependent way, a single scale F cannot absorb the shape difference and the extracted v2 will be biased. The paper's estimate that non-flow contributes less than 1% for pT>0.6 GeV/c comes from applying the template fit to PYTHIA 8, a generator without collective flow; this cannot validate the separation in real data where a flow component is also present. The paper's own observation that string-shoving PYTHIA produces negative v2 after the template fit shows that the extracted v2 is sensitive to the detailed shape and normalization of the template, which makes the concern concrete. I would like the authors to provide a closure test in which a known multiplicity- and species-dependent non-flow contribution is embedded in a flow-dominated sample and shown not to generate a spurious baryon-meson splitting, or to demonstrate the stability of the grouping and splitting with an alternative estimator such as four-particle cumulants.","section":"Section A.4, Eq. (A.1)"},{"comment":"The concluding claim that 'the system created in high-multiplicity p-Pb and pp collisions includes a stage with collectively flowing partons' is presented as the main physics conclusion, but the supporting model comparison with Hydro-Coal-Frag is only qualitative. The paper does not report a goodness-of-fit measure, nor does it quantify the level of agreement between the model and the measured v2 points, apart from visual inspection. More importantly, the claim of a '~5 sigma' splitting between baryons and mesons is not defined: it is not stated whether the significance is statistical only, whether systematic uncertainties are included, how the average over pT bins is performed, or which pairs of species are compared. Because the splitting is the key observable, the authors should give a precise definition of the significance and, if systematics are excluded, show explicitly how the significance changes when the systematic uncertainties from the template variation are included.","section":"Main text, Figs. 3--4 and conclusion"},{"comment":"The LM template is built from events with Nch<20, while the HM signal is measured at <Nch>~35, and the V0A/V0C multiplicity classes and FMD acceptances differ between pp and p-Pb. The paper should show, or at least summarize, the actual one-dimensional Delta-phi projections of the HM and LM correlation functions before and after the template scaling F, for at least one representative species and pT interval. Without such a visual or quantitative check, the reader cannot assess whether the assumed shape universality of the non-flow component is plausible, and whether the template fit is dominated by the flow term or by the F-scaled non-flow term at intermediate pT.","section":"Section A.1 and A.4"}],"minor_comments":[{"comment":"The caption of Fig. 2 repeats the phrase 'mesons (pi^pm, K^pm, K0S) and baryons (p+pbar, Lambda+Lambdabar)' in a way that is difficult to parse; please reformulate the caption to state once which symbols denote which species.","section":"Figure 2"},{"comment":"The combination v2^{PID} = sqrt(V2Delta^{TPC-FMD1,2} V2Delta^{TPC-FMD3} / V2Delta^{FMD1,2-FMD3}) should be accompanied by a brief statement of the underlying assumptions, in particular that the three V2Delta values share a common reference flow, otherwise the square-root formula is not transparent to the reader.","section":"Eq. (A.2)"},{"comment":"The phrase 'within ~1 sigma' for the grouping should be made quantitative: specify over which pT range and for which species pairs the compatibility is evaluated, noting that the pp measurement extends only to about 6 GeV/c while the p-Pb and Pb-Pb measurements reach 10 GeV/c.","section":"Abstract and introduction"},{"comment":"In the event-selection description, the pp HM class is given as '0.07% V0M' with Nch>25; the same quantity appears in the Fig. 2 caption as '0.07% V0M (-0' which appears truncated. Please correct the caption and ensure all multiplicity/centrality classes are printed consistently.","section":"Section A.1"},{"comment":"The comparison with the Hydro-Coal-Frag and Hydro-Frag models is shown with statistical uncertainties only; the text should state explicitly whether model uncertainties or parameter variations are included, since the claim that Hydro-Coal-Frag provides 'the best possible description' depends on the robustness of the model curves.","section":"Main text, Figs. 3--4"}],"recommendation":"major_revision","confidential_remarks":"This is a strong ALICE measurement that deserves publication after the template-fit robustness issue is addressed. The central claim is defensible but the load-bearing point is the validity of Eq. (A.1) against multiplicity-dependent non-flow; a closure test or an alternative estimator would substantially increase confidence. The abstract and conclusion use strong language ('demonstrate', 'evidence of a partonic flowing system') that should be moderated if the requested validation cannot be provided. I would not reject the paper, but I would not accept it in its current form."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Dear X,\n\nThe headline is real: this is the first observation of baryon-meson v2 grouping and ~5-sigma splitting at intermediate pT in high-multiplicity pp and p-Pb. That is a new and genuinely important measurement, especially because the CMS result cited as Ref. [24] did not see it. The analysis is careful, the event and track selections are standard ALICE, and the systematic tables are reasonably complete. The comparison to Hydro-Coal-Frag, with parameters tuned to spectra rather than to v2, is a fair and semi-independent test.\n\nNow the soft spots, in proportion. The load-bearing assumption is the template fit in Eq. A.1: Y_HM = F*Y_LM + flow. The LM template is Nch<20, the HM class is Nch~35, and the fit assumes the shape of non-flow scales by a single factor F. If per-particle jet or resonance correlations grow with multiplicity in a species-dependent way, the extracted v2 becomes biased and the bias can easily look like baryon-meson splitting. The paper's PYTHIA 8 test cannot close this gap, because PYTHIA has no collective flow by construction; it only shows that the method works when the template is correct. The paper itself notes that string-shoving PYTHIA gives negative v2 after the template fit, which is exactly the kind of sensitivity that worries me—it shows the result depends on the detailed shape of the LM template, not just on an overall flow scale. The ~5-sigma statement also appears to be statistical only; the systematic uncertainties are 2-8% and the template variation is evaluated within a fixed choice of LM class, not against a truly matched alternative.\n\nNone of this makes the measurement worthless. The grouping and splitting will stand as a datum, but the interpretation as evidence for a partonic flowing phase is stronger than the evidence. The paper would be much more convincing with a matched-template control, e.g., using multiplicity-binned LM sub-templates or repeating the extraction with different LM definitions and showing the splitting persists.\n\nWho is this for: the heavy-ion community and anyone arguing about QGP formation in small systems. It deserves a serious referee. I would send it to PRL or PLB with a request for a robustness check on the template subtraction. If that check holds, this becomes a very solid paper. If not, the partonic-flow claim is overstated.\n\nBest,\n[You]","headline":"First baryon-meson v2 grouping/splitting in small systems, but the partonic-flow conclusion leans on a template subtraction that the paper doesn't fully validate.","tokens_in":26851,"tokens_out":1623,"would_cite":true,"duration_ms":20410,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":["25.75.Ld","25.75.-q"],"model":"deepseek-v4-flash","headline":"Quark-level flow signature appears in proton-proton collisions","keywords":["elliptic flow","quark-gluon plasma","quark coalescence","small collision systems","proton-proton collisions","proton-lead collisions","partonic flow","two-particle correlations"],"falsifier":"A measurement of $v_2$ in the same high-multiplicity pp events using four-particle cumulants, which suppress non-flow, that does not show the baryon-meson grouping at $3 < p_T < 8$ GeV/c would undermine the claim; alternatively, a re-analysis with a multiplicity-dependent non-flow template from a controlled Monte Carlo without flow that removes the ~5$\\sigma$ splitting would falsify the extraction.","tokens_in":25745,"feed_emoji":"⚛️","tokens_out":8133,"duration_ms":72455,"temperature":0.7,"pith_summary":"This paper reports the first observation of a baryon-meson splitting in elliptic flow at intermediate transverse momentum in high-multiplicity proton-proton and proton-lead collisions. The pattern, in which the flow coefficients of mesons cluster together and those of baryons cluster at higher values, matches what is seen in heavy-ion collisions and is expected when flow develops among quarks before they combine into hadrons. If correct, the result implies that a stage of collectively flowing partons exists in small collision systems, not just in large nuclei.","feed_headline":"Quark-level flow signature appears in proton-proton collisions","feed_subtitle":"The baryon-meson flow split seen in heavy ions now appears in proton-proton and proton-lead collisions.","key_machinery":"The load-bearing mechanism is quark coalescence: when two or three flowing quarks combine to form a meson or baryon, the hadron inherits a $v_2$ determined by the summed $v_2$ of its constituent quarks, so baryons end up with a larger $v_2$ than mesons at the same $p_T$. The measurement technique is the long-range two-particle azimuthal correlation, with a large pseudorapidity gap between the correlated particles and a template fit that subtracts non-flow contributions using low-multiplicity events as a baseline.","core_discovery":"The paper's central claim is that the measured elliptic flow $v_2$ of identified hadrons in high-multiplicity p-Pb and pp collisions shows the same grouping by quark number as in Pb-Pb collisions: at transverse momenta of roughly 3 to 8 GeV/c, mesons ($\\pi^\\pm$, $K^\\pm$, $K^0_S$) share a common $v_2$ within about one standard deviation, baryons ($p+\\bar{p}$, $\\Lambda+\\bar{\\Lambda}$) share a higher $v_2$ within about one standard deviation, and the baryon-meson separation is about five standard deviations. This is the first time this pattern has been observed in small collision systems. The authors compare the data with a hybrid model that combines hydrodynamic flow of a quark-gluon plasma with quark-coalescence hadronization; this model reproduces the grouping and splitting, while a version without coalescence does not. They conclude that the systems created in high-multiplicity p-Pb and pp collisions include a stage with collectively flowing partons.","pith_inferences":["A natural extension would be to look for the same baryon-meson $v_2$ grouping among charm hadrons, such as $D$ mesons, in high-multiplicity proton-proton collisions, since number-of-constituent-quark scaling predicts a specific offset from light hadrons.","If the interpretation is correct, four-particle cumulant measurements, which are less sensitive to non-flow, should reproduce the splitting at intermediate $p_T$; a null result there would directly challenge the partonic-flow conclusion.","The result, if confirmed, implies that the event generators used as heavy-ion baselines should incorporate a partonic phase in small systems rather than relying on pure string fragmentation."],"forward_implications":["High-multiplicity proton-proton and proton-lead collisions at the LHC produce a system that contains a stage of collectively flowing partons, similar to heavy-ion collisions.","The observed grouping of meson $v_2$ values and baryon $v_2$ values at $p_T$ of 3–8 GeV/c is a new signature of quark-coalescence hadronization in small systems.","Models that include only hadronic rescattering or initial-state correlations fail to reproduce the baryon-meson splitting, so the data favor a partonic phase.","A hydrodynamic evolution followed by quark coalescence and fragmentation currently provides the best description of identified-hadron $v_2$ in both small and large collision systems."],"supporting_citations":[{"why":"Documents the baryon-meson $v_2$ grouping in Pb–Pb collisions that this paper seeks in small systems.","marker":"[14]"},{"why":"Introduces quark coalescence as the mechanism that translates quark-level flow into hadron-level flow.","marker":"[19]"},{"why":"Establishes the long-range two-particle correlation method for identified hadrons in p–Pb collisions.","marker":"[21]"},{"why":"Provides the template fit procedure used to subtract non-flow contributions.","marker":"[22]"},{"why":"Supplies the Pb–Pb comparison data and the particle-identification and $v_2$ extraction techniques.","marker":"[40]"},{"why":"Used to quantify the residual non-flow contamination of the template method via a pure non-flow simulation.","marker":"[43]"},{"why":"Calculations from the hydrodynamic-plus-quark-coalescence model that reproduce the observed grouping in p–Pb.","marker":"[47]"},{"why":"Calculations from the fragmentation-only variant that fails to reproduce the splitting, plus model curves for pp collisions.","marker":"[48]"}],"fun_headline_variants":["Quark flow seen in pp and p-Pb for the first time","First sign of partonic flow in proton-proton collisions","Baryon-meson split in small collisions hints at quark-gluon plasma","Proton-proton collisions show quark-level flow like heavy ions"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The analysis assumes that high-multiplicity two-particle correlations are exactly a scaled copy of low-multiplicity correlations (the non-flow template) plus a pure flow term, so if non-flow contributions change with multiplicity, the extracted flow coefficients and the baryon-meson split could be biased.","fun_headline_variants_meta":{"raw":{"variants":["Quark flow seen in pp and p-Pb for the first time","First sign of partonic flow in proton-proton collisions","Baryon-meson split in small collisions hints at quark-gluon plasma","Proton-proton collisions show quark-level flow like heavy ions"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000846,"raw_usage":{"total_tokens":3714,"prompt_tokens":1007,"completion_tokens":2707,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":623,"completion_tokens_details":{"reasoning_tokens":2631}},"tokens_in":623,"tokens_out":2707,"duration_ms":67575,"temperature":1.0,"reasoning_tokens":2631,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-12T20:45:36.014154+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"A measurement of $v_2$ in the same high-multiplicity pp events using four-particle cumulants, which suppress non-flow, that does not show the baryon-meson grouping at $3 < p_T < 8$ GeV/c would undermine the claim; alternatively, a re-analysis with a multiplicity-dependent non-flow template from a controlled Monte Carlo without flow that removes the ~5$\\sigma$ splitting would falsify the extraction.","supporting_citations":[],"review_version":1}