{"id":"0116fccd-08e0-4fa5-8736-11cfa70b58c1","arxiv_id":"2504.12161","paper_version":2,"verdict":"ACCEPT","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":4,"one_line_summary":"The first search for h->aa->mumu cc in LHC data sets 95% CL upper limits down to 3.3e-4 on sigma/sigma_SM times the branching fraction, excluding new parameter space in 2HDM+S models.","lead":"A search of CMS public 2016 data found no sign of the Higgs boson decaying into two new light particles that each decay to a muon pair and a charm quark pair. It sets the strongest limits yet in a previously unexplored beyond-standard-model scenario.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"CR/SR transfer of quarkonium resonance shapes and fixed family ratios is the main residual risk; a closure test with free SR ratios would settle it, but current evidence does not invalidate the search.","rationale":"The reader's weakest_assumption correctly identifies the CR-to-SR background transfer. I agree this is the most load-bearing assumption. However, the concern does not change the verdict. The jet-removal requirement (DeltaR<0.4 of any selected muon) means the c-tagged leading jet cannot be the jet from which the dimuon pair originates, so the tag is largely decorrelated from the m_mu+mu- observable. The background-only fit in the SR yields a good p-value (0.36), and the observed limits are consistent with the expected band with only minor local fluctuations, indicating no gross mis-modeling. The fixed family ratios are a conservative choice intended to prevent the five resonant background components from absorbing a single narrow signal peak; the Asimov signal-injection test reported in Section 6 shows no bias under the nominal background model. The proposed closure test — allowing the SR family ratios to float freely in the simultaneous fit and comparing limits — would provide a direct, data-driven validation of the transfer assumption and is a worthwhile addition, but its absence does not invalidate the published search. The paper is careful, reproducible via Open Data, and the limits are appropriately qualified. Therefore the verdict remains ACCEPT.","tokens_in":29636,"tokens_out":13469,"duration_ms":146598,"concrete_test":"Repeat the simultaneous maximum-likelihood fit of Section 8 using the public data, replacing the fixed SR family-ratio constraints with three free parameters (r2 = cSR_psi2S/cSR_Jpsi, r4 = cSR_Upsilon2S/cSR_Upsilon1S, r5 = cSR_Upsilon3S/cSR_Upsilon1S) and compare the best-fit values and the resulting 95% CL limits to the nominal fit. If the free ratios are consistent with the CR values within uncertainties and the limits shift by less than the expected band, the fixed-ratio transfer is benign; if not, the central limits are biased and require a modified background model.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The most load-bearing assumption is in Section 6: the SR background model fixes the within-family quarkonium ratios (cSR_psi2S/cSR_Jpsi, cSR_Upsilon2S/cSR_Upsilon1S, cSR_Upsilon3S/cSR_Upsilon1S) to the CR values and shares the resonance shape parameters {m0, sigma_L, sigma_R} between CR and SR. If the tight c-tag requirement in the SR alters the production composition of charmonium/bottomonium states or the dimuon mass resolution, the simultaneous fit could mis-model the SR background and bias the 95% CL limits, especially in the Upsilon mass region where signal and background peaks overlap. The paper argues the tag is decorrelated from the muon pair because jets within DeltaR<0.4 of a muon are removed, so the c-tagged jet is not the dimuon parent; this weakens but does not eliminate the risk, since the within-family ratios could still change if c-tagging selects different production mechanisms (e.g., feed-down contributions). The Asimov test described in Section 6 validates the model only under the assumption that the transferred ratios are correct, and the SR goodness-of-fit (p=0.36) is a necessary but not sufficient check. A direct closure test of the ratio transfer has not been presented.","agreement_with_reader":"agree"},"referee_report":null,"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Read this paper: it is the first LHC search for h→aa→μμcc, done by one person on 16.4 fb^-1 of CMS Open Data. The result is a credible set of 95% CL upper limits on σ/σ_SM × B(h→aa→μμcc), and the first constraints to cross unity in 2HDM+S Type II/III with tanβ=0.5 for ma between 4 and 11 GeV. That is genuinely new and closes a region that combined existing searches could not access.\n\nWhat the paper does well: the analysis is blinded; the background is estimated from data in a control region, with signal from independent Monte Carlo; the fit is validated with a goodness-of-fit p-value of 0.36; and the systematic treatment, including the low-pT muon efficiency and c-tagging uncertainties, is thorough. The author makes full use of Open Data, lists all datasets, and the work is reproducible. The use of DeepJet for a boosted, merged a→ccbar topology is a sensible first application.\n\nThe soft spot is in Section 6: the signal-region background model transfers the within-family quarkonium ratios (ψ(2S)/J/ψ, Υ(2S)/Υ(1S), Υ(3S)/Υ(1S)) from the CR and shares the resonance shape parameters {m0, σL, σR} between the two regions. If tight c-tagging changes the production composition (e.g., feed-down) or the dimuon mass resolution, the simultaneous fit could mis-model the SR and bias the limit. The author argues the tag is decorrelated from the muon pair, which is true as far as it goes, but a direct closure test with the SR ratios left free would settle it. This is the right thing for a referee to ask for. I do not think it invalidates the search: the assumption is reasonable, the CR fit is good, and the SR fit's p-value=0.36 is consistent. Still, it is an unquantified residual risk.\n\nMinor note: there are small local deviations near the 2σ level in the observed limits, consistent with background, and the paper says so.\n\nBottom line: this paper is for the exotic-Higgs-decay and 2HDM+S community, and for anyone interested in what a determined single analyzer can do with Open Data. It deserves a serious referee. I would send it out, and I would ask for the closure test. If that comes back clean, publish.","headline":"First search for h→aa→μμcc with CMS Open Data: credible limits that close an unconstrained 2HDM+S region; the CR/SR quarkonium-ratio transfer is the one residual worry.","tokens_in":30472,"tokens_out":3328,"would_cite":true,"duration_ms":30282,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":null,"created_at":"2026-08-16T12:36:31.597111+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":null,"supporting_citations":[],"review_version":1}