{"id":"38189e8f-2295-4d4f-9724-aa2ca0644bd6","arxiv_id":"2606.16258","paper_version":2,"verdict":"CONDITIONAL","confidence":"HIGH","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":7,"one_line_summary":"LEPS2 observes a 7.3σ localized enhancement below the K−pp threshold at low q in γd→K0Λp, with effective mass 2.354 GeV/c², supporting a K̄NN quasi-bound state in photoproduction.","lead":"A photoproduction experiment at SPring-8 reports a statistically significant low-momentum-transfer enhancement below the K−pp threshold in γd→K0Λp, interpreted as a K̄NN quasi-bound state. This is the first such claim in a photon-induced channel and complements earlier hadron-beam results.","discovery_kind":"extension","skeptic_critique":{"model":"grok-4.5","headline":"The 7.3σ claim rests on a signal template that forces a Voigt× Gaussian form factor; residual backgrounds or interference may still absorb the low-q excess without a quasi-bound state.","rationale":"The Reader correctly isolates the modeling assumption that underpins both the significance and the physical interpretation. The data show a clear localized excess; the analysis chain (exclusive selection, kinematic fit, 2-D templates, boundary-aware TS) is transparent and standard. The remaining soft spot is precisely that the signal template is highly constrained: any residual background or interference that also concentrates at low q will be absorbed by the Voigt× Gaussian component, inflating the apparent significance of a quasi-bound state. The paper already excludes one rescattering channel and checks NS stability, but does not close the more general class of smooth low-q mechanisms. That gap keeps the verdict CONDITIONAL rather than ACCEPT; no stronger downgrade is warranted because the excess itself is robust and the form-factor scale Q is consistent with E15. The concrete test above directly probes whether the BW pole is required or whether a non-resonant low-q amplitude suffices.","tokens_in":14234,"tokens_out":787,"duration_ms":6925,"concrete_test":"Re-fit the identical (M_Λp,q) histogram after replacing the Voigt× Gaussian signal template by a pure low-q reweighting of PS (no BW pole, only a free Gaussian or exponential form factor in q, optionally multiplied by a smooth threshold factor) and, separately, by an MC template of t-channel K* exchange + continuum interference. If either alternative yields ΔNLL comparable to the nominal TS = 64.9 and absorbs the subthreshold low-q excess, the quasi-bound-state interpretation (and the 7.3σ claim) is no longer uniquely supported.","verdict_should_be":"CONDITIONAL","load_bearing_attack":"The central claim (local 7.3σ support for a K̄NN quasi-bound state) is obtained from an extended ML fit of four templates (Sec. IV, Eq. 4–6). The signal is not free-form: it is the pure three-body phase-space template reweighted by a Voigt profile in M_Λp (BW mass M, width Γ, fixed resolution 11 MeV/c^{2}) times a Gaussian form factor exp(-q^{2}/Q^{2}). Backgrounds are QF (CLAS-weighted + AV18 Fermi), pure PS, and NS (sideband-reweighted PS with fixed Q_NS). The paper already notes that a two-step rescattering template fits to zero and that omitting NS leaves M,Γ,Q stable, yet it does not test whether a smooth, non-resonant amplitude that is itself peaked at low q (e.g., t-channel K* exchange interfering with continuum, or a coupled-channel cusp near the K⁻pp threshold) can produce a comparable localized excess once the same acceptance and kinematic-fit chain are applied. Because the likelihood-ratio TS = 64.9 and the Chernoff mixture (Eq. 7) are evaluated only against the background-only model that lacks any low-q peaking component, the quoted significance is conditional on the signal ansatz being the unique description of that peaking. The mass offset relative to E15 (2.354 vs 2.328 GeV/c^{2}) further suggests reaction-dependent distortion, so the same excess need not be a Breit–Wigner quasi-bound state.","agreement_with_reader":"agree"},"referee_report":{"model":"grok-4.5","summary":"The manuscript reports an exclusive measurement of γd → K⁰Λp with the LEPS2 solenoid spectrometer, searching for a K̄NN quasi-bound state. A localized excess is observed in the two-dimensional (M_Λp, q) distribution below the K⁻pp threshold and at low momentum transfer q. An extended maximum-likelihood template fit (quasi-free, phase-space, non-strange sideband, and a Voigt×Gaussian signal component) yields a local significance of 7.3σ under a Chernoff-type boundary correction. Effective shape parameters are extracted: M = 2.354 ± 0.011(stat.)^{+0.009}_{-0.005}(syst.) GeV/c², Γ = 0.055 ± 0.023(stat.)^{+0.031}_{-0.009}(syst.) GeV/c², Q = 0.350 ± 0.041(stat.)^{+0.035}_{-0.010}(syst.) GeV/c. The authors interpret the structure as supporting a K̄NN quasi-bound state in photoproduction, complementary to the J-PARC E15 result.","tokens_in":14676,"tokens_out":1324,"duration_ms":15110,"significance":"If the excess is indeed a K̄NN quasi-bound state, the result is significant: it constitutes the first claim of such a state in a photon-induced exclusive channel, provides an independent entrance channel (spin-triplet deuteron target, possible spin-flip amplitudes, t-channel K* exchange), and reports a form-factor scale Q consistent with E15. Strengths include the kinematic fit that improves M_Λp resolution to 11 MeV/c², full Geant4 acceptance embedding of templates, explicit checks that a two-step rescattering template fits to zero and that omitting the NS component leaves M, Γ, Q stable, systematic variations of vertex cuts and deuteron wave functions (AV18/Paris/CD-Bonn), and a properly boundary-aware significance evaluation. These elements make the observation a valuable addition to the still-inconclusive experimental landscape of kaonic nuclei.","major_comments":[{"comment":"Sec. IV and Eqs. (4)–(6): The quoted 7.3σ local significance and the claim of support for a quasi-bound state rest on a signal template that is phase-space reweighted by a Voigt profile in M_Λp times a Gaussian form factor exp(−q²/Q²). While the paper correctly shows that a dedicated two-step rescattering template yields a null contribution and that the NS component can be omitted without shifting the signal parameters, it does not test whether a smooth, non-resonant amplitude that itself peaks at low q (e.g., t-channel K* exchange interfering with continuum, or a coupled-channel distortion near the K⁻pp threshold) can absorb a comparable fraction of the localized excess once the same acceptance and kinematic-fit chain are applied. Because the likelihood-ratio TS = 64.9 is evaluated only against a background-only model lacking any low-q peaking component, the significance remains conditi","section":null},{"comment":"Sec. V.C (comparison with E15): The extracted mass lies ~26 MeV/c² higher than the E15 value (2.354 vs 2.328 GeV/c²) while Γ and Q are consistent within large uncertainties. The text correctly notes reaction-dependent distortions and possible interference, yet still presents the structure as evidence for the same K̄NN state. Given that the paper itself emphasizes that the parameters are “effective shape parameters” not directly comparable across channels, a clearer quantitative discussion of how large a mass shift is theoretically expected (or an explicit statement that the present data alone cannot establish identity with the E15 candidate) is required to avoid over-interpreting the agreement in Q.","section":null},{"comment":"Sec. V.A and Table 1: The fit reports N_Sig = 102 ± 30 against a total of roughly 700 events in the fit region, with QF still dominating (~60 %). The paper does not show the pull or residual distribution in the critical low-q, subthreshold corner after the background-only fit, nor does it quote the change in NLL when the Gaussian form-factor scale Q is fixed to a large value (i.e., no q peaking). These diagnostics would allow a reader to judge how much of the TS is driven by the localization in q versus the mass peak alone.","section":null}],"minor_comments":[{"comment":"Fig. 2 caption and projections: The shaded uncertainty band is mentioned but its construction (profile likelihood? Hessian?) is not stated; a one-sentence clarification would help.","section":null},{"comment":"Sec. III.A: The relative-likelihood cut R_p ≥ 0.1 and the Λ-pairing likelihood are described, yet no purity or efficiency numbers (from MC or sidebands) are given; a short table or sentence would strengthen the PID discussion.","section":null},{"comment":"Eq. (7) and the Chernoff mixture: The validation with 10⁵ pseudo-experiments is welcome; stating the highest TS observed in those toys would make the tail comparison fully transparent.","section":null},{"comment":"Throughout: Occasional notation inconsistencies appear (e.g., q vs. |p_Λp|, M vs. M_Λp); a uniform choice would improve readability.","section":null}],"recommendation":"major_revision","confidential_remarks":"The experimental work is careful and the observation of a localized excess is interesting. The main risk is over-interpretation of a model-dependent excess as definitive evidence for a quasi-bound state; the requested alternative-hypothesis tests should be feasible within the existing data set and would bring the paper to a solid publishable standard. Scope is appropriate for a letter-style nuclear-physics journal."},"author_rebuttal":null,"desk_editor":{"model":"grok-4.5","letter":"This is the first exclusive γd→K0Λp result that shows a clear low-q, subthreshold excess and claims a K̄NN quasi-bound state. That is genuinely new relative to the earlier inclusive LEPS search and complementary to J-PARC E15.\n\nWhat they did well is the experimental chain. Large-acceptance solenoid, PID likelihoods, displaced Λ vertex, kinematic fit that tightens M_Λp resolution to 11 MeV/c², and a binned 2-D extended ML fit with four templates (QF weighted to CLAS + AV18, pure PS, sideband NS, and signal). They check that a two-step rescattering template fits to zero, that dropping NS leaves M, Γ, Q stable, and that Fermi-wave-function swaps (Paris, CD-Bonn) do not invent the excess. The Chernoff mixture for the boundary at zero yield is the right asymptotic tool; 10^5 toys validate the tail. Forward K0 peaking is shown and matches the low-q kinematics they emphasize. Citations are appropriate and the E15 comparison is honest about the mass offset.\n\nThe soft spot is real but not fatal. The signal is not free-form: it is PS reweighted by a Voigt in M_Λp times a Gaussian form factor in q. The 7.3σ therefore measures improvement over a background model that has no low-q peaking component. They do not inject a smooth t-channel or coupled-channel continuum that itself prefers low q and re-fit; if that can absorb the excess, the significance drops. They already flag that M and Γ are effective, reaction-dependent shape parameters, not universal pole positions, and the mass sits ~26 MeV higher than E15. Spin-parity remains open. Those are the right caveats; they do not erase the localized excess in the data.\n\nThis is for people working on kaonic nuclei, strangeness in few-body systems, and photoproduction mechanisms. The analysis is transparent enough that a referee can demand the missing background-shape tests without starting from zero. I would send it to peer review; the result is important enough in the subfield and the execution is solid enough to deserve that scrutiny. Engage with it if you care about the K̄NN question; treat the quoted significance as conditional on the ansatz until the continuum alternatives are closed.","headline":"First exclusive photoproduction evidence for a K̄NN candidate: clean 2-D excess at low q, 7.3σ under a standard template, but the significance is conditional on the Voigt×Gaussian ansatz.","tokens_in":15481,"tokens_out":603,"would_cite":true,"duration_ms":5710,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"grok-4.5","headline":"Photoproduction of the deuteron shows a 7.3σ low-momentum-transfer excess below the K−pp threshold, supporting a K̄NN quasi-bound state.","keywords":["photoproduction","kaonic nuclei","K-bar NN quasi-bound state","gamma d to K0 Lambda p","momentum transfer","LEPS2","Breit-Wigner"],"falsifier":"A higher-statistics exclusive measurement of the same or a related photoproduction channel (for example with beam asymmetry or an isospin partner) that finds no localized low-q excess below the K−pp threshold after the same two-dimensional template fit would refute the claimed signal.","tokens_in":15130,"feed_emoji":"⚛️","tokens_out":929,"duration_ms":7058,"temperature":0.7,"pith_summary":"This experiment measures the exclusive γd → K^{0}Λp reaction with a large-acceptance solenoid spectrometer to look for the simplest kaonic nucleus, a quasi-bound K̄NN system. In the two-dimensional plane of Λp invariant mass and momentum transfer, a clear localized excess appears below the K−pp mass threshold and is concentrated at low momentum transfer. An extended maximum-likelihood template fit that includes quasi-free, phase-space, and non-strange backgrounds finds that a signal component with a Voigt mass line shape times a Gaussian form factor in momentum transfer is required at 7.3σ local significance. The fitted effective mass, width, and form-factor scale are reported; the form-factor scale is consistent with the earlier hadron-beam observation of a similar structure. The result supplies the first photoproduction evidence for a K̄NN quasi-bound state and shows that photon-induced reactions on the deuteron can form such compact systems at low momentum transfer.","feed_headline":"7.3σ excess points to a kaonic nucleus in photoproduction","feed_subtitle":"Low-momentum-transfer structure below the K−pp threshold seen in γd → K^{0}Λp at SPring-8","key_machinery":"The two-dimensional extended maximum-likelihood fit of the (M_Λp, q) histogram, in which a signal template is formed by reweighting three-body phase space with a Voigt (Breit–Wigner convolved with resolution) line shape in mass times a Gaussian form factor exp(−q^{2}/Q^{2}) in momentum transfer; the likelihood ratio of signal-plus-background versus background-only supplies the 7.3σ local significance.","core_discovery":"A localized enhancement is observed in the (M_Λp, q) distribution of γd → K^{0}Λp below the K−pp threshold and at low momentum transfer. A two-dimensional template fit yields a local significance of 7.3σ for a signal component whose effective shape parameters are M = 2.354 ± 0.011(stat.)+0.009/−0.005(syst.) GeV/c^{2}, Γ = 0.055 ± 0.023(stat.)+0.031/−0.009(syst.) GeV/c^{2}, and Q = 0.350 ± 0.041(stat.)+0.035/−0.010(syst.) GeV/c, supporting the existence of a K̄NN quasi-bound state in photoproduction.","pith_inferences":[],"forward_implications":[],"fun_headline_variants":["7.3σ low-q excess supports K̄NN quasi-bound state in γd→K⁰Λp","Localized M_Λp enhancement below K−pp threshold at 7.3σ local significance","LEPS2 sees 7.3σ signal for K̄NN state in photoproduction on deuteron","Low-momentum-transfer structure favors K̄NN quasi-bound state near threshold","Two-dimensional fit yields 7.3σ for K̄NN candidate with M=2.354 GeV/c²"],"cache_read_input_tokens":128,"weakest_assumption_plain":"The localized excess is assumed to be well described by a Breit–Wigner mass peak times a Gaussian form factor in momentum transfer, and residual backgrounds and interference are assumed not to generate a comparable low-momentum-transfer subthreshold structure.","fun_headline_variants_meta":{"raw":{"variants":["7.3σ low-q excess supports K̄NN quasi-bound state in γd→K⁰Λp","Localized M_Λp enhancement below K−pp threshold at 7.3σ local significance","LEPS2 sees 7.3σ signal for K̄NN state in photoproduction on deuteron","Low-momentum-transfer structure favors K̄NN quasi-bound state near threshold","Two-dimensional fit yields 7.3σ for K̄NN candidate with M=2.354 GeV/c²"]},"model":"grok-4.5","effort":"low","cost_usd":0.005796,"raw_usage":{"total_tokens":1700,"prompt_tokens":1005,"num_sources_used":0,"completion_tokens":116,"cost_in_usd_ticks":57960000,"prompt_tokens_details":{"text_tokens":1005,"audio_tokens":0,"image_tokens":0,"cached_tokens":256},"completion_tokens_details":{"audio_tokens":0,"reasoning_tokens":579,"accepted_prediction_tokens":0,"rejected_prediction_tokens":0}},"tokens_in":1005,"tokens_out":116,"duration_ms":4471,"temperature":1.0,"reasoning_tokens":579,"cache_read_input_tokens":256,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-07-14T17:50:46.866976+00:00","model_set":{"reader":"grok-4.5"},"falsifier":"A higher-statistics exclusive measurement of the same or a related photoproduction channel (for example with beam asymmetry or an isospin partner) that finds no localized low-q excess below the K−pp threshold after the same two-dimensional template fit would refute the claimed signal.","supporting_citations":[],"review_version":1}