{"id":"bbf3d148-0729-4c6f-968a-cfa88c4f739f","arxiv_id":"2412.09671","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":3,"one_line_summary":"In the Trajectum model, dilepton effective temperatures reliably track the quark-gluon plasma temperature, while photon effective temperatures are biased by flow and stay near 250-300 MeV.","lead":"This paper uses a heavy-ion collision simulation to show that dilepton pairs can measure the temperature of the quark-gluon plasma, while photons mostly cannot. It gives experimentalists a way to select dilepton masses and momenta to read off the plasma temperature at different times during the collision.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Early-time temperature claims depend on the poorly constrained hydrodynamic start time and on neglected pre-equilibrium emission, with the paper's own text identifying tau_hydro as the dominant uncertainty; a fixed-parameter rerun would settle the fragility.","rationale":"The reader's weakest-assumption analysis and my independent reading converge on the same load-bearing point: the early-time (<1 fm/c) temperature extraction is controlled by tau_hydro and by the absence of pre-equilibrium emission, both of which the paper explicitly identifies as dominant in the high-mass/high-pT region. This is not an internal inconsistency; the model-internal statement that dilepton Teff tracks the fluid temperature at later times is supported by Fig. 7, and the photon insensitivity claim is independently supported by the blue-shift mechanism and by the fixed-Tfluid fits in Fig. 5. The concern is external robustness: the experimental interpretability of the sub-1 fm/c temperatures and high-mass Teff values depends on assumptions that are known to be uncertain. The posterior sampling of 20 configurations already propagates tau_hydro uncertainty into the bands, but a targeted sensitivity scan at fixed extreme tau_hydro values clarifies whether the early-time points are controlled by tau_hydro or washed out by other parameters. Given that the authors themselves flag this limitation and the appendix partially mitigates the pre-equilibrium part, the reader's CONDITIONAL verdict is appropriate; my stress-test does not move it, so I recommend UNCHANGED.","tokens_in":18301,"tokens_out":6188,"duration_ms":70749,"concrete_test":"Re-run the 0-10% central collision dilepton spectra with the MAP parameters but with tau_hydro fixed to the lower and upper 1-sigma values (0.21 and 0.55 fm/c), re-extract Teff and average emission times for the 2-3 and 3-4 GeV/c^2 mass bins, and compare with the current systematic bands; if the shifts exceed those bands, the early-time mapping is not robust. As a complementary check, append the pre-equilibrium dilepton production of Garcia-Montero et al. to the thermal spectrum and re-extract Teff in the same bins; if the combined Teff moves by more than the systematic uncertainty, the omission is material for the high-mass claims.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central claim that dilepton Teff tracks the QGP fluid temperature is model-internal and well supported for emission times above roughly 1 fm/c, but the headline early-time extraction (average times below 1.0 fm/c, Table I) is loaded onto precisely the two ingredients the paper itself flags as weakest. Section II reports tau_hydro = 0.38 ± 0.17 fm/c, and Section III states that 'the dominant parameter contributing to this uncertainty is the starting time of hydrodynamics tau_hyd'. Section VI adds that pre-equilibrium and Drell-Yan contributions 'start to play an important role at higher invariant masses'. Since the early-time bins are the high invariant-mass bins (2-3 and 3-4 GeV/c^2) with average emission times of about 1.4 and 0.9 fm/c, a shift of tau_hydro within its 1-sigma range changes both the emitting volume before 1 fm/c and the relative weight of hot early fluid cells; adding pre-equilibrium emission would harden the high-mass spectrum and push Teff upward. The appendix comparison to Garcia-Montero et al. mitigates the pre-equilibrium concern by showing their thermal+pre-equilibrium curve lies within the quoted systematic uncertainty, but that comparison starts their thermal emission at tau = 1.0 fm/c, so it does not validate the sub-1 fm/c mapping. The fragility is therefore specific to the early-time and high-mass claims, not to the later-time Teff-vs-temperature correspondence shown in Fig. 7.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The authors extend the Trajectum hybrid model with next-to-leading-order thermal photon and dilepton emission rates from a weakly coupled QGP, use 20 posterior parameter samples to estimate systematic uncertainties, and compute transverse-momentum and invariant-mass spectra, effective temperatures, average emission times, and elliptic flow for Pb-Pb collisions at sqrt(s_NN)=5.02 TeV over several centrality classes. They find that the photon T_eff is approximately 250-300 MeV and largely insensitive to the true fluid temperature because of radial-flow blue-shift, while the dilepton T_eff extracted from the intermediate-mass-region spectrum tracks the fluid temperature. They further show that selecting on dilepton mass and transverse momentum selects average emission times from late emission (about 5.6 fm/c) down to below 1.0 fm/c, and that dilepton elliptic flow can be mapped as a function of emission time. The appendix compares the photon spectra and flow with ALICE data and with previous theoretical calculations, and compares the dilepton results with those from Kasmaei and Strickland and from Garcia-Montero et al.","tokens_in":18582,"tokens_out":9476,"duration_ms":96330,"significance":"If the conclusions hold, this is a useful model-based study: it provides a physics interpretation of measured effective temperatures, argues that thermal dileptons are better thermometers than thermal photons, and demonstrates how mass and transverse-momentum selections can map the temperature and flow history of the quark-gluon plasma. The manuscript has clear strengths: it is built on a Bayesian-constrained dynamical model, it propagates posterior-based uncertainties, it compares with ALICE data and several prior calculations, and it states explicit caveats about omitted pre-equilibrium and Drell-Yan contributions. The main weakness is that the most novel part of the claim, namely access to average emission times below 1 fm/c, depends on the weakly constrained hydrodynamic starting time tau_hydro and on the absence of pre-equilibrium emission, both of which the authors themselves flag as the dominant early-time uncertainty.","major_comments":[{"comment":"The headline result that dilepton mass selection gives average emission times below 1 fm/c is carried by exactly the two ingredients the paper identifies as least reliable. Section II quotes tau_hydro = 0.38 +/- 0.17 fm/c from the Bayesian fit, and Section III states that the dominant parameter contributing to the early-time uncertainty is tau_hydro because pre-hydrodynamic photon and dilepton production is not implemented. The 3 < m_ll < 4 GeV/c^2 bins in Table I have mean emission times of 0.83-0.96 fm/c, only a few tenths of a fm/c after the central tau_hydro; within one sigma tau_hydro ranges from 0.21 to 0.55 fm/c, which changes both the amount of hot fluid available before 1 fm/c and the temperature assigned to the earliest fluid cells. The appendix comparison with Garcia-Montero et al. shows their thermal+pre-equilibrium curve within the Trajectum systematic band, but their thermal emission starts at tau = 1.0 fm/c, so it does not validate the sub-1 fm/c mapping. I therefore request a sensitivity study that varies tau_hydro over its posterior (or fixes it to several values) and, if possible, an estimate of pre-equilibrium dilepton emission in the high-mass windows before the paper claims extraction of fluid temperatures at average times below about 1 fm/c.","section":"Section III, Table I, Fig. 7"},{"comment":"The relation T_eff approximately T_fluid for dileptons is established in Fig. 5 by generating the spectrum from rates that depend on T and then fitting that spectrum with Eqs. (1)-(2); the agreement is therefore a calibration of the extraction procedure inside the model, not an independent test of the relationship in the real QGP. The manuscript should state this limitation explicitly and, where possible, confront the prediction with LHC dilepton spectra or an independent rate implementation rather than only with other thermal-model calculations and the ALICE photon spectrum. This does not undermine the later-time part of the analysis, but it sets the appropriate strength of the claim that thermal dileptons are much better probes of the QGP temperature.","section":"Section IV, Fig. 5"}],"minor_comments":[{"comment":"The text says the table shows average emission times specifically for 0-10% and 30-40% collision centralities, but Table I contains only 0-10% columns; either add the 30-40% values or correct the sentence.","section":"Section III, Table I"},{"comment":"Reference [27] is incomplete: it gives only '(2024), arXiv:2402.01998 [nucl-ex]' without author or title; please complete the citation.","section":"References"},{"comment":"The text contains several instances of the missing space in 'theTrajectum' (e.g., in the abstract and at the start of Section II); please fix these formatting issues throughout.","section":"Abstract and Section II"},{"comment":"The sentence 'we have showcased in Fig. 6 that by applying different selection criteria on the transverse momentum and the invariant mass it is possible to discriminate between different average emission times' refers to a figure that shows effective temperatures versus centrality; the average-emission-time discrimination is actually presented in Table I and Fig. 7, so the cross-reference should be corrected.","section":"Section VI"}],"recommendation":"major_revision","confidential_remarks":"The early-time claim is the most newsworthy part and also the most fragile; a sensitivity analysis varying tau_hydro and/or adding pre-equilibrium emission should be feasible with the existing framework. I would not block publication on the model-internal calibration point, but the wording of the central claim should be moderated accordingly."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"This is a careful, useful paper. The central result holds up within the model: dilepton effective temperatures extracted from invariant mass in the 1–3 GeV window track the average fluid temperature, while thermal photon Teff stays near 250–300 MeV regardless of the true temperature because radial flow blue-shifts the pT spectrum. That is a clean demonstration, and it is what people will cite.\n\nWhat is actually new: the first implementation of NLO thermal photon and dilepton rates in Trajectum, with systematic uncertainties propagated from the Bayesian posterior, a comprehensive Teff map across centralities and selection windows, and a showing that dilepton mass and pT selections pick out average emission times from about 5.6 fm/c down to below 1 fm/c. The flow section is a nice addition, especially the time-selected v2 that shows high-mass dileptons carry early-flow information.\n\nThe authors are honest about their limitations. They state plainly that tau_hydro = 0.38 ± 0.17 fm/c is poorly constrained and dominates the systematic uncertainty of early emission, and they note that pre-equilibrium and Drell-Yan contributions become important at high invariant mass. That is where the real soft spot sits. The headline early-time extraction, average times below 1 fm/c, comes from the 2–3 and 3–4 GeV/c^2 mass bins. Those bins are exactly where an incorrect tau_hydro shifts the emitting volume before 1 fm/c, and where missing pre-equilibrium emission would harden the spectrum and push Teff up. The appendix comparison to Garcia-Montero et al. helps for masses up to 4 GeV, but their thermal emission starts at 1 fm/c, so it does not validate the sub-1 fm/c mapping. A rerun with a fixed larger tau_hydro, or an estimate of the pre-equilibrium contribution in those mass windows, would settle how much the early-time numbers move.\n\nI also agree with the reader that the statement about agreement with ALICE direct photon Teff is a bit stronger than the evidence: the model comparison to ALICE in the appendix is for the spectrum, not for the extracted Teff, and the experimental Teff includes prompt photons and other contributions. That overstatement is minor, but easy to fix.\n\nOn the circularity point: yes, the Teff–Tfluid relation is calibrated from spectra generated by rates that depend on T, so it is not an independent test. But the paper frames it as a calibration study, and that is legitimate. The model is constrained to bulk data, so the predictions for these EM observables are not free.\n\nWho is this for? Anyone working on electromagnetic probes of the QGP, and the experimental community planning future dilepton measurements. It deserves a serious referee. I would send it to review, with the request that the authors address the early-time fragility quantitatively — either by showing tau_hydro sensitivity or by including pre-equilibrium estimates in the relevant mass windows.","headline":"Solid Trajectum calibration study showing dilepton Teff tracks fluid temperature while photon Teff is flow-dominated; the sub-1 fm/c early-time claims are the genuinely fragile part.","tokens_in":19150,"tokens_out":1417,"would_cite":true,"duration_ms":16647,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":["25.75.-q","12.38.Mh"],"model":"deepseek-v4-flash","headline":"Dilepton invariant-mass spectra track the true quark–gluon plasma temperature, while thermal photon effective temperatures are dominated by radial flow.","keywords":["quark-gluon plasma","effective temperature","thermal photons","thermal dileptons","heavy-ion collisions","hydrodynamics","elliptic flow","Trajectum"],"falsifier":"Repeat the same Trajectum calculation with $\\tau_{\\rm hydro}$ varied over its Bayesian uncertainty, $0.38 \\pm 0.17$ fm/c, and with pre-equilibrium emission included; if the extracted effective temperature in the 2–3 GeV/c² mass window moves by more than the quoted systematic band, the claimed early-time thermometer would fail. A precise measurement of the dilepton invariant-mass spectrum in that window at collider energies that disagrees with the model's predicted $T_{\\rm eff}$ would also falsify the correspondence.","tokens_in":18075,"feed_emoji":"🔥","tokens_out":11435,"duration_ms":99575,"temperature":0.7,"pith_summary":"This paper uses the Trajectum heavy-ion model, constrained by a Bayesian fit to Pb–Pb collisions at 5.02 TeV, to ask what thermal photons and dileptons actually measure about the quark–gluon plasma temperature. Its central result is that the effective temperature read from thermal dilepton invariant-mass spectra in the 1–3 GeV/c² range tracks the true fluid temperature, while the effective temperature read from thermal photon transverse-momentum spectra stays near 250–300 MeV, almost independent of centrality, because radial flow blue-shifts the photons. Cutting on dilepton mass and transverse momentum selects average emission times from about 5.6 fm/c down to below 1.0 fm/c, turning dileptons into a stage-by-stage thermometer of the plasma. The same selections map the elliptic flow of the system over its lifetime. Because electromagnetic radiation is the only probe that escapes the plasma without rescattering, knowing which measured slope is a real temperature and which is a flow artifact is essential for interpreting heavy-ion experiments.","feed_headline":"Dileptons track the QGP temperature; photons just see flow","feed_subtitle":"Selection on dilepton mass and momentum maps QGP temperatures from late times to under 1 fm/c.","key_machinery":"The load-bearing machinery is the event-by-event implementation of next-to-leading-order thermal emission rates for photons and dileptons in the Trajectum hybrid model, which evolves a generalized initial state through a pre-equilibrium stage, 2+1D viscous hydrodynamics, and a hadronic afterburner. Photon and dilepton yields are sampled per fluid cell from the NLO production rates, built on the electromagnetic current-current correlator; the hadron-resonance-gas contribution to photons is added through the afterburner. Effective temperatures are extracted by fitting the Boltzmann form $\\exp(-p_T/T_{\\rm eff})$ for photons and the $(m_{ll} T_{\\rm eff})^{3/2} \\exp(-m_{ll}/T_{\\rm eff})$ form for dileptons in the intermediate mass region $1<m_{ll}<3$ GeV/c². The identity that carries the argument is that the dilepton invariant mass is Lorentz invariant, so the slope of $dN/dm_{ll}$ is not blue-shifted by flow, while the photon $p_T$ is; this is what lets $m_{ll}$ selections act as a clock for the fluid temperature.","core_discovery":"The paper claims that, within the Trajectum framework, the effective temperature extracted from the thermal dilepton invariant-mass distribution in the intermediate mass region is a faithful thermometer for the quark–gluon plasma, whereas the photon effective temperature is not. Because the dilepton pair mass is Lorentz invariant, the extracted $T_{\\rm eff}$ in each mass bin equals the average temperature of the fluid cells that dominate that bin, and this correspondence holds across centralities: the $T_{\\rm eff}$ points lie on the model's own $\\langle T\\rangle(\\tau)$ curve when plotted at the median emission time. Thermal photons, by contrast, have their transverse momentum blue-shifted by radial flow, so their $T_{\\rm eff}$ saturates near 250–300 MeV regardless of the actual plasma temperature; observing a photon temperature above the crossover therefore does not prove the plasma was that hot. Using $p_T$ and $m_{ll}$ selections, the model resolves emission-time windows from late ($\\langle \\tau \\rangle \\approx 5.6$ fm/c) to very early ($\\langle \\tau \\rangle < 1.0$ fm/c), and the corresponding dilepton elliptic flow maps how flow builds up over time, with high-mass pairs showing near-zero $v_2$ because they come from the earliest stages.","pith_inferences":["Because the model's $T_{\\rm eff}(m_{ll})$ curve sits on its own $\\langle T\\rangle(\\tau)$ curve, comparing the same extraction with experimental dilepton spectra would directly test the hydrodynamic temperature profile; a mismatch would point to the starting time or equation of state rather than to the emission rates.","The sub-1 fm/c temperatures are the least robust part of the map: they depend on $\\tau_{\\rm hydro}$ and ignore pre-equilibrium emission, so the early-time values are likely to shift once those contributions are included.","The centrality independence of the photon plateau suggests the same 250–300 MeV effective temperature would appear at lower collision energies, meaning historical photon-based temperature estimates may partly measure flow rather than heat.","A two-dimensional selection in $m_{ll}$ and $p_T$ could sharpen the emission-time resolution further and turn the dilepton spectrum into a tomographic cooling curve; this extension is not worked out in the paper."],"forward_implications":["A measured dilepton spectrum in the intermediate mass region gives direct access to the quark–gluon plasma temperature evolution, not just a single integrated number.","Photon-only effective temperatures near 300 MeV should not be quoted as evidence that the plasma reached a particular temperature; the same plateau can come from a cooler fluid with radial flow.","Selections on dilepton invariant mass and transverse momentum can map both temperature and elliptic flow as functions of emission time, from about 5.6 fm/c to below 1 fm/c.","High-mass, high-$p_T$ dileptons are sensitive to the earliest, least-constrained stage of the collision, so precise measurements there can tighten the parameters that govern the initial state.","The hadron-resonance-gas contribution to photons matters mainly at low $p_T$; separating it from the QGP contribution is needed before using photon slopes as temperature indicators."],"supporting_citations":[{"why":"Supplies the next-to-leading-order thermal photon production rate used for the QGP photon emission in Trajectum.","marker":"[66]"},{"why":"Supplies the next-to-leading-order thermal dilepton production rate used for the QGP dilepton emission.","marker":"[67]"},{"why":"Earlier model calculation showing dilepton effective temperatures track the QGP temperature, which this paper confirms in its own framework.","marker":"[42]"},{"why":"Companion calculation supporting the claim that dileptons are a good probe of QGP temperature and early-time emission.","marker":"[43]"},{"why":"Bayesian fit to Pb–Pb data that constrains the Trajectum parameters, including the hydrodynamic starting time, and supplies the posterior samples used for the systematic bands.","marker":"[58]"},{"why":"Introduces the Trajectum hybrid model whose initial conditions, hydrodynamics, and hadronization carry the whole simulation.","marker":"[50]"},{"why":"Earlier hybrid-model photon calculation with prompt photons, used as the main theoretical comparison for the photon spectrum.","marker":"[45]"},{"why":"Experimental direct photon spectrum used as the baseline comparison for the thermal photon yields.","marker":"[22]"},{"why":"Pre-equilibrium electromagnetic emission calculation that the paper identifies as the way to reduce the dominant early-time systematic from tau_hydro.","marker":"[46]"}],"fun_headline_variants":["Dileptons are the QGP's true thermometer; photons only probe flow","Thermal photons fail to reveal QGP heat; dileptons do it right","Dilepton mass bins pick out QGP eras from late to <1 fm/c","Dilepton flow exposes QGP's time evolution stage by stage"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The paper's early-time temperatures assume hydrodynamics starts near 0.38 fm/c and that nothing radiates before that time; if the starting time is wrong or pre-equilibrium emission is sizeable in the mass and $p_T$ windows used, the sub-1 fm/c temperatures and the high-mass $T_{\\rm eff}$ values would shift.","fun_headline_variants_meta":{"raw":{"variants":["Dileptons are the QGP's true thermometer; photons only probe flow","Thermal photons fail to reveal QGP heat; dileptons do it right","Dilepton mass bins pick out QGP eras from late to <1 fm/c","Dilepton flow exposes QGP's time evolution stage by stage"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.00141,"raw_usage":{"total_tokens":5784,"prompt_tokens":1120,"completion_tokens":4664,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":736,"completion_tokens_details":{"reasoning_tokens":4579}},"tokens_in":736,"tokens_out":4664,"duration_ms":31561,"temperature":1.0,"reasoning_tokens":4579,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-11T16:51:09.801776+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Repeat the same Trajectum calculation with $\\tau_{\\rm hydro}$ varied over its Bayesian uncertainty, $0.38 \\pm 0.17$ fm/c, and with pre-equilibrium emission included; if the extracted effective temperature in the 2–3 GeV/c² mass window moves by more than the quoted systematic band, the claimed early-time thermometer would fail. A precise measurement of the dilepton invariant-mass spectrum in that window at collider energies that disagrees with the model's predicted $T_{\\rm eff}$ would also falsify the correspondence.","supporting_citations":[{"cited_title":"Holographic Photon Production in Heavy Ion Collisions","cited_arxiv_id":"1609.07208","evidence_quote":"Supplies the next-to-leading-order thermal photon production rate used for the QGP photon emission in Trajectum."},{"cited_title":"smash-transport/smash: Smash- 3.1,","cited_arxiv_id":null,"evidence_quote":"Bayesian fit to Pb–Pb data that constrains the Trajectum parameters, including the hydrodynamic starting time, and supplies the posterior samples used for the systematic bands."}],"review_version":1}