{"id":"d65f4913-1c85-46d6-986e-d5f2a111a892","arxiv_id":"2510.27174","paper_version":2,"verdict":"ACCEPT","confidence":"MODERATE","novelty_score":8.0,"correctness_risk":"low","formal_verification":"none","parameter_count":0,"one_line_summary":"First observation of D_s0(2317)+ → D_s*+ γ with >10σ significance; its branching ratio relative to D_s+π0 is measured as 7.14%.","lead":"Physicists at Belle and Belle II have observed, for the first time, a rare radiative decay of the D_s0(2317)+ meson, in which it emits a photon and a D_s*+ meson. The measured rate, presented as a ratio to its main decay, provides a new experimental handle on whether this puzzling meson is a quark–antiquark state, a hadron molecule, or a mixture.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Missing D_s*+→D_s+γ branching-fraction correction in the efficiency formula may bias the reported R by ~6.5%.","rationale":"The reader identified MC modeling of efficiency and the broken-signal background as the weakest assumption, but the more load-bearing and easily checkable issue is the treatment of B(D_s*+→D_s+γ). This intermediate branching fraction appears only through the MC chain; the paper gives no indication that the 93.5% radiative decay probability of the D_s*+ has been divided out. The reported R is what feeds the physics interpretation and the comparison with theory. A ~6.5% relative bias is larger than the total systematic uncertainty and comparable to the statistical uncertainty, so it is not a cosmetic detail. The observation of the decay itself would not be invalidated, so this is not a rejection; rather, the paper should be accepted only after the authors confirm/clarify the D_s*+ branching-fraction convention and, if needed, correct the central value and update the theory plot. I agree with the reader that the analysis is internally coherent and the 10σ observation is convincing; my concern is specifically about the quantitative ratio claim.","tokens_in":14960,"tokens_out":16957,"duration_ms":165328,"concrete_test":"Inspect the decay table used to generate the D_s*+ in the signal MC. If D_s*+ is forced to D_s+γ with 100% probability (as is typical for dedicated signal samples), recompute R by dividing the quoted value by B(D_s*+→D_s+γ) = 0.935 and check whether the central value shifts outside the quoted uncertainty. If the decay table already assigns the PDG branching fractions to D_s*+ modes and ε_exp counts accepted/generated D_s0→D_s*γ events, then verify the ε_exp definition and confirm no correction is needed. Either way, state explicitly how B(D_s*+→D_s+γ) enters the efficiency or formula.","verdict_should_be":"CONDITIONAL","load_bearing_attack":"The central quantified claim is the branching-fraction ratio R = B(D_s0(2317)+→D_s*+γ)/B(D_s0(2317)+→D_s+π0) = [7.14 ± 0.70 ± 0.23]%. The extraction uses N_exp(D_s*+γ) = R × N^fit_exp(D_s+π0) × ε_exp(D_s*+γ)/ε_exp(D_s+π0). The text states the signal MC simulates D_s0(2317)+→D_s*+γ with phase space and then “the decay D_s*+ → D_s+ γ is simulated as a P-wave decay,” but nowhere does the paper divide by the known branching fraction B(D_s*+→D_s+γ), nor is it included in the systematic table. If the dedicated signal MC forces D_s*+→D_s+γ with 100% probability, then ε_exp(D_s*+γ) is the selection efficiency for that forced chain and the fitted R is actually B(D_s0→D_s*γ)×B(D_s*→D_sγ)/B(D_s0→D_s+π0). Using the PDG value B(D_s*+→D_s+γ) ≈ 93.5%, the true ratio would be 7.14%/0.935 ≈ 7.64%, a ~6.5% relative shift—far larger than the quoted 3.2% total systematic uncertainty. The observation of the radiative decay at 10σ would survive, but the paper’s headline branching-fraction measurement and the Fig. 3 comparison with theory would be biased. If instead the MC generated D_s*+ with PDG branching fractions and ε is defined as accepted events per generated D_s0→D_s*γ decay, no correction is needed; however, the manuscript does not state this, leaving the central number ambiguous and potentially mislabeled.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper reports the first observation of the radiative decay D_s0(2317)+ -> D_s*+ gamma using Belle (980.4 fb^-1) and Belle II (427.9 fb^-1) data. A blinded simultaneous unbinned extended maximum-likelihood fit to the M(D_s*+ gamma) spectra yields a signal significance of 10.1 sigma and a branching-fraction ratio R = B(D_s0(2317)+ -> D_s*+ gamma)/B(D_s0(2317)+ -> D_s+ pi0) = [7.14 +/- 0.70(stat.) +/- 0.23(syst.)]%. Separate fits to each dataset are consistent, and the result is compared with theoretical predictions for molecular and quark-antiquark interpretations of the D_s0(2317)+.","tokens_in":15383,"tokens_out":17157,"duration_ms":168082,"significance":"If the quoted ratio is correct, this is the first quantitative measurement of this radiative transition and a new experimental constraint on the internal structure of the D_s0(2317)+. The analysis is blinded, uses MC-validated fit models, and the existence of the decay at high significance is not in doubt. However, the central branching-fraction ratio and the comparison in Fig. 3 depend on a normalization detail concerning the D_s*+ -> D_s+ gamma decay that the manuscript does not currently document unambiguously. The observation claim is robust, but the headline number needs clarification or correction.","major_comments":[{"comment":"The extraction formula near Fig. 2 sets N_exp(D_s*+ gamma) = R N^fit_exp(D_s+ pi0) e_exp(D_s*+ gamma)/e_exp(D_s+ pi0), with R defined as B(D_s0->D_s*gamma)/B(D_s0->D_s pi0). The MC description says the D_s*+ -> D_s+ gamma decay is simulated as a P-wave decay, but no factor for the physical branching fraction B(D_s*+ -> D_s+ gamma) appears anywhere, including in the systematic table. If the signal MC forces this decay, then e_exp(D_s*+gamma) is conditional on that forced chain and the fitted R is actually B(D_s0->D_s*gamma)*B(D_s*->D_s gamma)/B(D_s0->D_s pi0). Using PDG B(D_s*->D_s gamma) ~93.5%, the true ratio would be 7.14%/0.935 ~7.64%, a ~6.5% shift, much larger than the quoted 3.2% total systematic. The authors must either demonstrate that their MC includes the physical D_s* branching fraction (so that e_exp is defined per generated D_s0->D_s*gamma decay including the D_s* decay prob","section":"Data and simulated MC samples; simultaneous fit to M(D_s*+ gamma)"},{"comment":"The reference-channel yields are quoted as N^fit_exp(D_s+ pi0) = 10820 +/- 230 (Belle) and 6108 +/- 163 (Belle II). These statistical uncertainties are not propagated into the quoted statistical uncertainty on R: the simultaneous fit fixes N^fit_exp(D_s+ pi0) to their point estimates, and the pseudo-experiments in the systematic section fluctuate the reference yields only by their systematic uncertainties. The 2-3% statistical uncertainties of the reference yields should be included, e.g., as Gaussian-constrained nuisance parameters or added in quadrature to the statistical error. This raises sigma_stat(R) from 0.70 to about 0.71; the effect is numerically small but the reported error budget is formally incomplete.","section":"Fit to M(D_s+ pi0); simultaneous fit"}],"minor_comments":[{"comment":"The central value is quoted as 7.13% in the abstract, 7.14% in the body, and 7.13% in the Fig. 3 caption. Please unify.","section":"Abstract vs. body and Fig. 3"},{"comment":"The extracted caption text 'Exp: 0.74+-7.13' appears to be a formatting error; it should read 'Exp: 7.13 +/- 0.74'.","section":"Fig. 3 caption"},{"comment":"There is a typo 'backgrond' in the paragraph following Fig. 1.","section":"General"},{"comment":"The systematic table lists 'xp reweighting' and 'MC sample size' with entries only in one column. Please clarify whether these uncertainties apply to both channels or only to the D_s* gamma channel; the quadrature sum suggests the latter.","section":"Table I"},{"comment":"The signal MC uses a phase-space model for D_s0(2317)+ -> D_s*+ gamma. If the true radiative amplitude has a non-isotropic angular distribution, the detection efficiency could be biased. Please justify this modeling choice or assign a corresponding systematic uncertainty.","section":"Data and simulated MC samples"}],"recommendation":"major_revision","confidential_remarks":"To the editor: The observation of the decay is convincing and the paper is generally well executed, but the central branching-fraction ratio depends on an undocumented normalization point: whether the signal MC forces D_s*+ -> D_s+ gamma or includes the physical D_s* branching fraction. If forced, the quoted R is biased by ~6.5%. This must be resolved before acceptance. If the authors confirm that the MC includes the physical branching fraction, the issue reduces to clarity and the paper could be accepted after minor revision; if not, the central number and theory comparison must be corrected."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Here's my take on arXiv:2510.27174. The headline result—first observation of D_s0(2317)+ → D*+ γ, with a 10.1σ significance in the combined Belle/Belle II data—looks solid. The signal is visible in both detectors, the analysis is blinded, the fit is sensible, and the systematic budget covers the usual suspects. This is a genuinely new measurement and a useful input for the spectroscopy of the D_s0(2317)+.\n\nThe soft spot is the normalization of the branching-fraction ratio. The paper defines R = B(D_s0→D*γ)/B(D_s0→D_sπ0) and reports [7.14 ± 0.70 ± 0.23]%. But I don't see anywhere a division by B(D*+ → D_s+ γ), which is about 93.5%. The signal MC description says the D*+ → D_s+ γ decay is simulated as a P-wave; it does not say what branching fraction is used for that sub-decay. If the dedicated signal MC forces the D* to decay to D_sγ with 100% probability, then the efficiency ε(D*γ) is for the full chain, and the fitted R is actually R × B(D*→D_sγ). That would shift the central value from 7.14% to about 7.64%, nearly 7% relative—much larger than the 3.2% total systematic uncertainty. The 10σ observation survives, but the comparison with theory in Fig. 3 and the discussion about molecular vs. c̄s assignments would be biased.\n\nI don't want to overstate this—it's possible the MC generator uses the physical B(D*→D_sγ) in the efficiency and the cancellation is already built in. But the paper should say so explicitly. The current text leaves the central number ambiguous. That's the main issue I'd raise.\n\nOther things are minor. The abstract and the text give 7.13 vs. 7.14 for R. The broken-signal and xp-reweighting uncertainties are handled with MC studies; they could be off, but the paper quantifies them and the fit is stable. The separate Belle and Belle II results are consistent.\n\nOverall, this is a careful, important experimental result. It deserves a serious referee, and the review should ask the authors to pin down the D* branching-fraction handling before publication. If the correction is needed, the number changes but the physics conclusion—that the radiative mode is observed and the ratio is larger than most molecular predictions—probably doesn't.","headline":"First observation is solid at 10σ; but the quoted R may be missing the B(D*+→D_s+γ) factor, so the central number needs a clarification or a correction.","tokens_in":17928,"tokens_out":6246,"would_cite":true,"duration_ms":53741,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"deepseek-v4-flash","headline":"10-sigma signal reveals D_s0(2317)+ radiative decay","keywords":["D_s0(2317)+","radiative decay","branching fraction ratio","charmed mesons","exotic hadrons","molecular state","quark-antiquark state","e+e− collider"],"falsifier":"Measure the same ratio using the alternative decay D_s*+ → D_s+π0 instead of D_s*+ → D_s+γ to reconstruct the D_s*+. Because the two D_s*+ decay modes have very different photon energies, a consistent ratio would validate the photon-efficiency corrections; a large discrepancy would indicate the reweighting or broken-signal modeling is wrong. Alternatively, a future dataset that measures the ratio with comparable precision and finds a value outside 7.14% ± systematic uncertainties would falsify the central claim.","tokens_in":14888,"feed_emoji":"⚛️","tokens_out":5501,"duration_ms":46779,"temperature":0.7,"pith_summary":"This paper reports the first observation of the radiative decay of the charmed-strange meson D_s0(2317)+ into a D_s*+ meson and a photon. The signal is seen in continuum e+e−→cc̄ data with a significance above 10 standard deviations. The branching-fraction ratio to the known D_s+π0 mode is measured as 7.14±0.70±0.23%. Because the D_s0(2317)+ sits below the DK threshold and its internal quark structure is debated, this ratio is a sensitive discriminator between a conventional quark-antiquark state and a loose molecule. The value is higher than molecular-model predictions and lower than some quark-model estimates, pointing toward a mixed or a specific quark-model interpretation.","feed_headline":"10-sigma signal reveals D_s0(2317)+ radiative decay","feed_subtitle":"Branching ratio to D_s*+γ is 7.14%, splitting the gap between molecule and quark-model predictions.","key_machinery":"The analysis is carried by the ratio R = B(D_s0(2317)+ → D_s*+γ)/B(D_s0(2317)+ → D_s+π0), extracted from a simultaneous unbinned extended maximum-likelihood fit to the D_s*+γ invariant-mass spectra from the two experiments. The detection efficiencies are obtained from Monte Carlo samples that are reweighted by the measured xp (reduced-momentum) distribution of the reference hadronic channel, and the small 'broken-signal' peaking background—a real D_s+ paired with a random photon—is fixed from simulation at 7.5% (9.3%) of the signal in the two samples.","core_discovery":"The central claim is that the radiative decay D_s0(2317)+ → D_s*+γ exists and occurs at a rate (7.14±0.70±0.23)% of the hadronic D_s+π0 decay. The paper establishes this by reconstructing D_s+ → K+K−π+, combining with a photon to form D_s*+, then adding another photon to form the D_s0(2317)+ candidate, and fitting the M(D_s*+γ) spectrum in a blind analysis of two data samples. The simultaneous fit yields signal yields of about 712 and 387 events in the two experiments, with a combined significance of 10.1σ, treating the systematic uncertainty as a Gaussian smearing. This is the first quantitative measurement of this mode, moving the field from upper limits to a non-zero branching fraction.","pith_inferences":["If the ratio is confirmed, it would imply the radiative width of D_s0(2317)+ is substantial, which could be cross-checked by measuring the absolute width once the D_s+π0 branching fraction becomes known.","A natural extension is to measure the same ratio using the D_s*+ → D_s+π0 decay chain instead of the photon chain; agreement would validate the efficiency corrections for low-energy photons.","The xp-reweighting technique could become a template for branching-fraction-ratio measurements in other charm spectra where signal Monte Carlo is limited."],"forward_implications":["The measured ratio of about 7% rules out the simplest molecular-only explanations, which predict 0.5–4.25%, as the sole nature of D_s0(2317)+.","It also disfavors the larger pure-quark-model expectation above 8.1% from some models, but agrees with light-front and chiral quark model predictions for a c̄s state.","The measurement gives a concrete target for mixed c̄s-molecular models to match, potentially pinning the admixture.","It demonstrates that rare radiative modes of the D_s0(2317)+ can be observed, opening the same technique for D_s1(2460)+ radiative decays.","As a first measurement, it provides a normalization point for theoretical calculations of radiative widths of exotic mesons."],"fun_headline_variants":["10σ signal: D_s0(2317)+ → D_s*+ γ finally observed","First radiative decay of D_s0(2317)+: 7.14% ratio at 10σ","D_s0(2317)+ radiates? Yes, 10σ evidence for γ emission","Belle combo finds D_s0(2317)+ radiative decay at 10σ","D_s0(2317)+ → D_s*+ γ: now measured, not just limits"],"cache_read_input_tokens":2304,"weakest_assumption_plain":"The result rests on the Monte Carlo modeling of the signal detection efficiency—specifically the reweighting of the signal simulation by the xp distribution measured in the D_s+π0 channel—and on the fixed fraction of 'broken-signal' background (a real D_s+ combined with a random photon) taken from simulation; if either is biased, the central ratio shifts, though the 10σ signal would likely persist.","fun_headline_variants_meta":{"raw":{"variants":["10σ signal: D_s0(2317)+ → D_s*+ γ finally observed","First radiative decay of D_s0(2317)+: 7.14% ratio at 10σ","D_s0(2317)+ radiates? Yes, 10σ evidence for γ emission","Belle combo finds D_s0(2317)+ radiative decay at 10σ","D_s0(2317)+ → D_s*+ γ: now measured, not just limits"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000288,"raw_usage":{"total_tokens":1550,"prompt_tokens":793,"completion_tokens":757,"prompt_tokens_details":{"cached_tokens":256},"prompt_cache_hit_tokens":256,"prompt_cache_miss_tokens":537,"completion_tokens_details":{"reasoning_tokens":634}},"tokens_in":537,"tokens_out":757,"duration_ms":46533,"temperature":1.0,"reasoning_tokens":634,"cache_read_input_tokens":256,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-04T07:00:52.928676+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Measure the same ratio using the alternative decay D_s*+ → D_s+π0 instead of D_s*+ → D_s+γ to reconstruct the D_s*+. Because the two D_s*+ decay modes have very different photon energies, a consistent ratio would validate the photon-efficiency corrections; a large discrepancy would indicate the reweighting or broken-signal modeling is wrong. Alternatively, a future dataset that measures the ratio with comparable precision and finds a value outside 7.14% ± systematic uncertainties would falsify the central claim.","supporting_citations":[],"review_version":1}