{"id":"f270481a-e84b-44f0-8d9e-6ac5bddc2705","arxiv_id":"2507.18900","paper_version":2,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":5.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":3,"one_line_summary":"Reanalysis of KEK-PS E325 data finds a velocity-independent low-mass excess in rho/omega dielectron spectra, interpreted as a ~10% downward mass shift of rho and omega mesons in nuclear matter without significant broadening.","lead":"This paper presents updated measurements of rho and omega meson masses in 12 GeV proton-nucleus collisions, split into three velocity bins. A low-mass excess near the omega peak appears in every bin for carbon and copper targets, consistent with a roughly 10% in-medium mass reduction with no significant width broadening.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"k1 extraction rests on unvaried production-point geometry; quoted errors omit this model systematic.","rationale":"The reader's weakest-assumption identification is correct: the production-point distribution is the sharpest place where an untested modeling choice enters the conversion from a robust data excess to a quantitative mass shift. The paper itself is transparent about this assumption and notes that JAM production points were not used because JAM places mesons outside the nuclear radius, but the replacement model is ad hoc and unvaried. The low in-medium fractions in Table 15 make the k1 extraction especially sensitive to path-length prescriptions: small changes in the number of mesons that decay while traversing high-density material translate directly into the amplitude of the shifted spectral component. The quoted systematic errors in Table 18 are extensive for experimental effects, but they omit this geometric model uncertainty and also omit the related choice of asymmetric in-medium spectral shape. In that sense the concern is not that the excess is spurious or that the mass shift is zero; it is that the precision of the headline k1 values overstates what is actually constrained. This matches the reader's CONDITIONAL verdict, so I recommend no change to the verdict rather than a new one. Agreement is 'agree' because the reader identified the same load-bearing assumption in its weakest_assumption field. A concrete re-fit with alternative production-point distributions is the right check because it directly quantifies whether k1 moves outside the currently quoted errors, and if it does not, the conditional verdict could later be upgraded.","tokens_in":45780,"tokens_out":13348,"duration_ms":160496,"concrete_test":"Re-run the case (ii) fits for C and Cu with identical JAM momenta and data, but replace the incident-side surface-hemisphere production points with two alternatives: (a) production points distributed throughout the full target volume weighted by the Woods-Saxon density, and (b) production points uniformly distributed on the full spherical surface at the half-density radius. Compare the fitted k1 in each beta-gamma bin against the Table 18 systematic errors. If the resulting k1 shifts by more than the quoted +0.01/-0.09 systematic in the middle bin, or by more than the stated statistical errors elsewhere, the quoted precision does not cover this model uncertainty and the central numeric claim should be made explicitly conditional on the production-point distribution.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The measurement side is solid: the low-mass excess is large, reproducible across targets and beta-gamma bins, and the experimental systematics are extensively evaluated. The load-bearing weakness is the geometric assumption used to convert that excess into a mass shift. In Section 6.1, 'it is assumed that the production points of the rho and omega mesons are uniformly distributed on the surface of the incident-side hemisphere at the half-density of the target nucleus,' an assumption inherited from Ref. [20] that is not varied anywhere in the systematic budget. Table 15 shows that the in-medium decay fractions for omega are small at k2=0: 8.6% for C and 14% for Cu in the lowest beta-gamma bin, falling to 2.3% and 4.1% in the highest bin. A volume-distributed production point, or even a full-surface distribution, would shorten the average in-medium path length and change the amount of shifted signal required to match the same excess, directly changing the fitted k1. The systematic errors quoted in Table 18 cover fit region, bin width, mass scale, mass resolution, event mixing, and electron-ID efficiency, but they do not include production-point geometry, JAM momentum uncertainty, or the choice among in-medium spectral shapes such as nBW/m^3 versus mdRBW2. The direction is likely toward larger k1 if shorter in-medium paths are assumed, so the existence of a nonzero mass shift is probably robust; however, the central quantitative statement of roughly a 10-12% shift at normal nuclear density is not yet bounded by a systematic that is demonstrably part of the model uncertainty.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper reports an updated analysis of the KEK-PS E325 data for 12 GeV p+C and p+Cu reactions, measuring e+e- invariant mass spectra in the rho/omega region and splitting them into three beta-gamma bins. The authors find a significant low-mass excess relative to known hadronic sources in all six target/beta-gamma combinations (Sec. 5.2, Table 13; 8.5-13.4 sigma). They interpret this excess with a Monte Carlo model in which the rho and omega pole masses scale as 1 - k1 rho/rho0 and the widths scale as 1 + k2 rho/rho0, with the mesons produced on the incident-side hemisphere at half nuclear density (Sec. 6.1). For the asymmetric nBW/m^3 line shape (case (ii)), they obtain k1 ~ 0.10-0.12, corresponding to about a 10% downward mass shift at normal nuclear density, with no significant width broadening. The paper emphasizes that this is the first beta-gamma-differential extraction from E325 and compares the result with the earlier E325 analysis, with CLAS and TAPS, and with QCD sum-rule predictions.","tokens_in":46114,"tokens_out":2429,"duration_ms":29021,"significance":"If the mass-shift interpretation is correct, this is an important cold-nuclear-matter measurement of vector-meson mass modification with a new velocity dependence, in a regime where previous experiments disagree. The measurement side is genuinely strong: the excess is large, reproduced across two targets and three beta-gamma bins, the analysis uses the full statistics, and the experimental systematics for the excess ratio are extensive (Sec. 5.3, Table 14). The paper also ships an unusually transparent account of detector calibration, simulation tuning, background construction, and model assumptions, which supports reproducibility of the analysis. The main significance is therefore the robust existence of a low-mass excess and its beta-gamma dependence; the conversion of that excess into a specific ~10% mass shift is model-dependent rather than parameter-free, and the paper itself acknowledges that one of its three model cases fails to reproduce the known pp rho/omega ratio. The result is a credible and important update, but the quantitative 'mass modification' claim should be treated as an interpretation under stated assumptions.","major_comments":[{"comment":"The extracted k1 depends directly on the assumed production-point distribution, stated in Sec. 6.1 as uniform production 'on the surface of the incident-side hemisphere at the half-density of the target nucleus' and inherited from Ref. [20]. This assumption is never varied in the systematic-error evaluation: Table 18 lists fit region, bin width, mass scale, mass resolution, event-mixing, and electron-ID efficiency, but not the production geometry. Since Table 15 shows in-medium decay fractions that differ strongly between C and Cu and between beta-gamma bins (e.g., 8.6% vs 14% for omega in the lowest bin), a volume-distributed or full-surface production would change the path-length distribution and therefore the fitted k1. The authors should either propagate this model uncertainty into the quoted errors or demonstrate that the extracted k1 is insensitive to it.","section":"Sec. 6.1 and Table 18"},{"comment":"The paper itself states that case (i) (symmetric nBW shape) yields rho/omega ratios of 0.02-0.25 for C and <0.14 for Cu, contradicting the pp value of 1.0 +/- 0.2 from Ref. [61], and concludes that 'the simulated model must be further refined to accurately reproduce the data.' This is a self-identified failure of one of the three model variants used to extract the central result. Because Table 16 shows that k1 changes from 0.12-0.16 in case (i) to 0.10-0.12 in cases (ii) and (iii), the quantitative mass shift is not robust across the modeling choices considered. The paper should either exclude the failing case from the central claim, quantify the resulting model spread, or explicitly present the final k1 values as conditional on the asymmetric-shape assumption.","section":"Sec. 6.2, case (i) and Table 17"},{"comment":"The systematic errors quoted for k1 (e.g., +0.01/-0.03 for beta-gamma <2.1) cover only the detector- and background-related variations listed in A-F. They do not include the uncertainty in the JAM momentum distributions, the choice among the in-medium spectral shapes (nBW, nBW/m^3, mdRBW2, or the previous-study shape shown in Fig. A1), or the two assumed branching-ratio prescriptions (constant branching ratio vs. constant partial width). Each of these can shift the extracted k1 by an amount comparable to or larger than the statistical errors in Table 16. The quoted 'mass decrease of about 10%' therefore understates the total model uncertainty, and the authors should widen the systematic budget or present the result with a clear model-uncertainty caveat.","section":"Sec. 6.2.1 and Table 18"}],"minor_comments":[{"comment":"There are several typographical slips, including 'Backgound' in Sec. 4.8, 'Collaoration' in Ref. [2], and an incomplete decay label 'rho->e+e' in Sec. 2.1; these should be corrected in a final proofreading pass.","section":"Throughout"},{"comment":"The excess ratio is defined as N_excess/(N_omega + N_excess), but this definition is only stated in the text after Table 13 and would be clearer if repeated in the table caption, since the reader needs it to interpret Fig. 32.","section":"Sec. 5.2, Table 13"},{"comment":"The parameter k1 is identified with the mass shift 'at normal nuclear density' in the conclusion, but Eq. (11) defines it via M(rho)/M(0) = 1 - k1 rho/rho0, so k1 is the density-slope parameter; the paper should consistently call k1 the fractional shift per unit density and reserve 'mass decrease at rho0' for the combination k1 rho(r)/rho0, which is only equal to k1 at rho = rho0.","section":"Sec. 6.1, Eq. (11)-(12)"},{"comment":"The iterative weighting method for the combinatorial background is described clearly, but the convergence criterion is not stated; adding a sentence on when the iterative loop is stopped would improve reproducibility.","section":"Sec. 4.8.3"}],"recommendation":"major_revision","confidential_remarks":"The measurement of the low-mass excess and its beta-gamma dependence is solid and deserves publication after revision. My main reservation is that the quantitative k1 extraction inherits unvaried modeling assumptions, and one of the paper's own model cases is inconsistent with the known pp rho/omega ratio. The authors should be asked to either include the production-geometry and line-shape choices in the systematic budget or soften the central quantitative claim. I would not reject the paper on these grounds, but the revision needs to address the model-uncertainty issue rather than only polishing the text."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Punchline first: the low-mass excess in the rho/omega region is real, and this paper is the cleanest statement of the E325 result so far. The 10% mass shift, however, is a model-dependent number whose error bar does not include the two assumptions that move it most.\n\nWhat is new: first beta-gamma-resolved rho/omega mass spectra from pA, doubling of omega statistics (6900 to 14860), optimized acceptance, PHOTOS internal radiative corrections, Pluto Dalitz form factors, and the CLAS-style asymmetric BW/m^3 line shape. The excess is present in all three velocity bins and both targets at 8.5-13.4 sigma, and the fits excluding the excess region give chi2 probabilities 0.016-0.61. I trust the data excess. The experimental systematics (mass scale, resolution, event mixing, electron ID) are carefully evaluated, and the authors do not bury the tensions with CLAS/TAPS.\n\nThe soft spots are all in the conversion from excess to mass shift. k1 and k2 are free parameters fitted to the very spectra that show the excess, and in case (i) the symmetric Breit-Wigner fit drives rho/omega to about zero, contradicting the pp value of 1.0±0.2 at the same beam momentum; the authors themselves say that model must be refined. The adopted case (ii) with BW/m^3 gives k1 about 0.10-0.12 and k2~0, but the production-point distribution is assumed to be uniform on the incident-side hemisphere at half density (Sec. 6.1), inherited from Naruki 2006 and never varied in Table 18. The in-medium decay fractions for omega are small (2-14%), so the fitted shift is extremely sensitive to where the mesons are produced. JAM momentum uncertainty and the choice of in-medium mass shape are also not in the systematic budget. The direction of the missing systematics is likely toward larger k1 if in-medium path lengths are shorter, so I would not conclude the shift is an artifact; I would conclude its size is not yet bounded by the quoted errors.\n\nThis paper is for everyone working on in-medium vector mesons; it is the strongest current statement of the mass-shift side against the CLAS/TAPS broadening picture, and the authors handle that comparison honestly by testing an omega-unmodified variant and discussing reaction-mechanism differences.\n\nRecommendation: send it to peer review, but the referees should demand either a production-geometry systematic (volume vs. surface, full vs. hemisphere) and a JAM momentum variation, or an explicit argument why those choices cannot change k1 within the quoted errors. Without that, the central 'about 10%' claim overstates the evidence.","headline":"The low-mass excess is robust and the beta-gamma-resolved spectra are a real step forward, but the 10% mass shift rests on unvaried production-geometry and line-shape assumptions that need to be part of the systematic budget.","tokens_in":46805,"tokens_out":2798,"would_cite":true,"duration_ms":31389,"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":"Rho and omega mesons produced in 12 GeV p+C and p+Cu collisions show a low-mass excess consistent with a roughly 10% in-medium mass drop at normal nuclear density.","keywords":["vector meson mass modification","in-medium hadron properties","dielectron invariant mass","rho meson","omega meson","chiral symmetry restoration","nuclear matter density","proton-nucleus collisions"],"falsifier":"Measure the same reaction on a deuterium or hydrogen target with the same detector: if the low-mass excess and the fitted $k_1$ vanish, the effect is genuinely nuclear; if they persist, the excess has a non-nuclear origin.","tokens_in":45572,"feed_emoji":"⚛️","tokens_out":8022,"duration_ms":81515,"temperature":0.7,"pith_summary":"By reanalyzing the full dielectron dataset of the E325 experiment, this paper tries to establish that rho and omega mesons produced in 12 GeV proton collisions with carbon and copper nuclei are lighter inside the nucleus. In all three meson-velocity bins and for both targets, the e+e- invariant-mass spectrum shows an excess on the low-mass side of the omega peak that known hadronic sources cannot reproduce; the excess is 8.5 to 13.4 standard deviations. A model in which the pole mass falls linearly with nuclear density, $M(\\rho) = (1 - k_1 \\rho/\\rho_0) M(0)$, reproduces the spectra with $k_1$ between 0.10 and 0.12, i.e., a roughly 10% mass reduction at normal nuclear density. If right, this is direct evidence for density-dependent hadron masses in cold nuclear matter and supports the idea that hadron mass is tied to chiral symmetry breaking in the QCD vacuum. The paper also provides the first velocity ($\\beta\\gamma$) dependence of the modification, which stays nearly flat across the measured range.","feed_headline":"Rho and omega mesons lose ~10% of mass inside nuclei","feed_subtitle":"Low-mass excesses at 8.5-13.4 sigma in dielectron spectra point to density-dependent vector-meson mass shifts.","key_machinery":"The mechanism that carries the argument is the density-dependent pole mass and width inserted into a stepwise decay simulation: $M(\\rho) = (1 - k_1 \\rho/\\rho_0) M(0)$ and $\\Gamma(\\rho) = (1 + k_2 \\rho/\\rho_0) \\Gamma(0)$, with the Woods-Saxon density profile of the target. The key observable is the excess ratio, the number of counts in the 0.62-0.76 GeV/$c^2$ region beyond known sources relative to the omega yield, and the fitted parameter $k_1$ that converts that excess into a mass shift. The velocity dependence enters because lower-$\\beta\\gamma$ mesons decay a larger fraction inside the nucleus (for the Cu target, rho in-medium decay fractions range from 76% at $\\beta\\gamma<2.1$ down to 48% at $\\beta\\gamma>2.7$), so the same density-dependent shift produces a stronger distortion in slow-meson spectra. The asymmetric line shape $\\mathrm{nBW}/m^3$, previously used in gamma-nucleus analyses, is what allows the rho contribution to coexist with the mass-shifted omega.","core_discovery":"The central discovery is that the low-mass excess survives a much more careful analysis than the original observation: with more than twice the omega statistics, optimized acceptance, updated Dalitz form factors, internal radiative corrections, and a re-tuned detector simulation, the excess appears in every $\\beta\\gamma$ bin for both C and Cu, with significances from 8.5$\\sigma$ to 13.4$\\sigma$. Fitting the excess with a density-dependent pole mass yields $k_1 = 0.12^{+0.03}_{-0.03}\\,\\mathrm{(stat.)}^{+0.01}_{-0.03}\\,\\mathrm{(sys.)}$ for $\\beta\\gamma<2.1$, $0.12^{+0.04}_{-0.06}\\,\\mathrm{(stat.)}^{+0.01}_{-0.09}\\,\\mathrm{(sys.)}$ for $2.1<\\beta\\gamma<2.7$, and $0.10^{+0.03}_{-0.05}\\,\\mathrm{(stat.)}^{+0.02}_{-0.02}\\,\\mathrm{(sys.)}$ for $\\beta\\gamma>2.7$, corresponding to a roughly 10% mass drop at normal nuclear density. A symmetric Breit-Wigner shape forces the fitted rho yield to zero, while an asymmetric $\\mathrm{nBW}/m^3$ shape restores $\\rho/\\omega$ ratios consistent with proton-proton measurements; the paper therefore adopts the asymmetric shape and finds no significant width broadening ($k_2$ consistent with zero). The authors present this as updated evidence that the in-medium effect in cold nuclear matter is a mass shift rather than broadening.","pith_inferences":["A direct test of the geometric assumption would be a target-thickness scan: if the fitted $k_1$ changed systematically with target thickness, the assumed production-point distribution rather than the mass shift would be carrying the excess.","Extending the same fitting framework to the phi meson in the same three $\\beta\\gamma$ bins could map the mass shift as a function of strangeness content; the earlier phi result suggests a smaller shift, and a common analysis would put the rho/omega and phi modifications on the same footing.","A hydrogen or deuterium target run with identical acceptance would measure the vacuum line shape and $\\rho/\\omega$ ratio in the same detector, breaking the degeneracy between an unmodified broadened rho and a mass-shifted omega.","The flat $\\beta\\gamma$ dependence implies that at even higher meson momenta the observed excess should shrink as the in-medium decay fraction falls; future higher-energy experiments can verify this trend and turn the velocity dependence into a measurement of the meson's in-medium lifetime."],"forward_implications":["If the claim is right, the rho and omega pole masses drop by about 10% at normal nuclear density in cold matter, moving from roughly 775 and 782 MeV/$c^2$ toward about 700 MeV/$c^2$.","Because $k_1$ stays nearly constant across the three $\\beta\\gamma$ bins, the mass shift in the sampled momentum range has little momentum dependence, which can be compared directly with QCD sum-rule calculations of the momentum dependence.","The data rule out a pure broadening interpretation: fits that allow only width growth cannot reproduce the excess, so any future model must include a downward density-dependent mass shift.","The necessity of the asymmetric $\\mathrm{nBW}/m^3$ shape means the in-medium resonance is not a simple symmetric peak, a constraint that effective-model calculations of vector-meson spectral functions must satisfy."],"supporting_citations":[{"why":"Previous version of this analysis whose production-point model and mass-shift interpretation the present work updates.","marker":"[20]"},{"why":"QCD sum-rule predictions of roughly 16% mass reduction at normal density, the theoretical benchmark compared with the fitted $k_1$.","marker":"[6, 7]"},{"why":"Gamma-nucleus measurement that supplies the asymmetric $\\mathrm{nBW}/m^3$ line shape adopted here and the no-mass-shift result being compared.","marker":"[34]"},{"why":"Hadronic cascade model that provides the generated meson momentum distributions used in the decay-path Monte Carlo.","marker":"[42]"},{"why":"Event generator used for the Dalitz decay invariant-mass shapes and updated form factors.","marker":"[43]"},{"why":"Radiative-correction code applied to the resonance decay mass spectra before detector simulation.","marker":"[40]"},{"why":"Proton-proton measurement giving the $\\rho/\\omega \\simeq 1.0$ ratio used as a sanity check on the fitted yields.","marker":"[61]"}],"fun_headline_variants":["Velocity-dependent mass shift seen for rho and omega in p+A","Vector mesons shed mass inside nuclei, velocity-dependent","Rho and omega mass drop ~10% in nuclei, no broadening","Mass shift, not broadening, explains vector meson excess in p+A","In-medium vector mesons: mass drops, width holds"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The load-bearing premise is the assumed starting point of each meson: to convert the spectrum into a mass shift, the model places every rho and omega uniformly on the incident-side surface of the target nucleus where the density is half its central value, and the paper does not vary this assumption when assigning systematic errors.","fun_headline_variants_meta":{"raw":{"variants":["Velocity-dependent mass shift seen for rho and omega in p+A","Vector mesons shed mass inside nuclei, velocity-dependent","Rho and omega mass drop ~10% in nuclei, no broadening","Mass shift, not broadening, explains vector meson excess in p+A","In-medium vector mesons: mass drops, width holds"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000792,"raw_usage":{"total_tokens":3533,"prompt_tokens":1032,"completion_tokens":2501,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":648,"completion_tokens_details":{"reasoning_tokens":2413}},"tokens_in":648,"tokens_out":2501,"duration_ms":17632,"temperature":1.0,"reasoning_tokens":2413,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-15T18:06:06.462857+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Measure the same reaction on a deuterium or hydrogen target with the same detector: if the low-mass excess and the fitted $k_1$ vanish, the effect is genuinely nuclear; if they persist, the excess has a non-nuclear origin.","supporting_citations":[{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Gamma-nucleus measurement that supplies the asymmetric $\\mathrm{nBW}/m^3$ line shape adopted here and the no-mass-shift result being compared."},{"cited_title":"Blobel et al","cited_arxiv_id":null,"evidence_quote":"Proton-proton measurement giving the $\\rho/\\omega \\simeq 1.0$ ratio used as a sanity check on the fitted yields."}],"review_version":1}