{"id":"bdfca239-ce5c-4994-84a3-a173ca13cf79","arxiv_id":"2506.18956","paper_version":1,"verdict":"CONDITIONAL","confidence":"LOW","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":4,"one_line_summary":"MicroBooNE observes no significant excess of neutral-current Delta radiative decay single photons, which disfavors enhanced-Delta explanations of the MiniBooNE anomaly but leaves zero-proton single-photon sources unconstrained.","lead":"This doctoral thesis reports MicroBooNE's search for single photons from neutrino-induced Delta baryon radiative decay, the largest expected single-photon source in the detector, used to test photon-like explanations of the MiniBooNE low-energy excess. The measured events match expectation, but extra single photons without any visible proton activity cannot be ruled out.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The 0p single-photon prediction is underconstrained by data; the stated consistency with expectation depends on simulated 0p photon rates and reconstruction efficiencies that the thesis itself shows are not settled.","rationale":"The reader's weakest_assumption identifies the same load-bearing point: the null result is only as strong as the simulated prediction for single photons with no visible proton activity, the topology the thesis concedes it cannot constrain. My independent reading reaches the same conclusion from the in-scope text. The abstract states the central claim with an explicit caveat: data are consistent with nominal expectation, but 0p sources cannot be ruled out. That claim is internally consistent. However, the quantitative 'consistent' statement depends on 0p photon production and reconstruction modeling. The thesis itself provides direct evidence that this modeling is not settled: Sec. 5.3.1 describes Geant4 EM model modifications; Figs. 4.39-4.43 document a GENIE branching-ratio bugfix and differences between GENIE FSI models for single-photon production; Fig. 4.35 shows true-0p efficiencies that are simulation-derived; and Sec. 4.6.5 introduces x_0p because data cannot determine the 0p fraction. These are not external speculations but limitations stated in the manuscript, so per the reviewing rule they should be weighed in the verdict. The visible-proton exclusions are not attacked by this concern; they rest on topologies where the proton activity is directly reconstructed and checked against sidebands. The reader's verdict of CONDITIONAL already captures the need to inspect the truncated Sec. 4.6 results and to treat the 0p modeling caveat seriously. My stress-test does not move that verdict; it reinforces the condition. No code or data artifacts are shipped, and this is a thesis preprint rather than a collaboration-reviewed publication, so independent reproducibility is limited; that supports keeping the conditional framing. A concrete computational check using alternate FSI and EM models would settle whether the 0p simulation uncertainty is large enough to change the exclusion conclusions.","tokens_in":59114,"tokens_out":3428,"duration_ms":40341,"concrete_test":"Recompute the Sec. 4.6.5 two-dimensional (x_Delta, x_0p) exclusion and the Sec. 4.6.4 one-bin LEE test using (a) an alternate GENIE FSI model from the set compared in Figs. 4.42-4.43 and (b) the modified Geant4 EM model described in Sec. 5.3.1, with all other selections and systematics fixed. If the confidence level or excluded region for the 0p-only hypothesis (x_0p near 1) changes by more than the quoted one-sigma systematic band, then the 0p simulation, not merely the caveat, is the decisive ingredient in the null result.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central claim is explicitly a null result: data are consistent with the nominal GENIE-plus-detector expectation, with the 0p (no visible proton) topology left as the admitted gap. That consistency statement is only as strong as the simulated 0p prediction. Two ingredients carry the load here. First, the predicted 0p photon background depends on GENIE's NC Delta radiative photon rate and on photonuclear absorption and final-state interactions; Figs. 4.39-4.43 document a GENIE branching-ratio bugfix and material differences among GENIE FSI models, while Sec. 5.3.1 documents ad hoc Geant4 EM model modifications. The thesis itself therefore demonstrates that these models are not settled. Second, the acceptance for true-0p single-photon events is a simulation-based efficiency (Fig. 4.35; Secs. 5.2.2-5.2.3), and the 0p fraction x_0p is introduced in Sec. 4.6.5 precisely because data cannot pin it down. If the true 0p efficiency is lower than simulated, or if true photonuclear absorption is higher, a genuine 0p photon excess could be absorbed into the 'consistent with expectation' conclusion; if the reverse, the 0p exclusion could be overstated. This concern does not weaken the abstract's explicit caveat, which already concedes that 0p sources cannot be ruled out, but it means the quantitative content of 'consistent' is not robust for the 0p channel. The visible-proton exclusion is not affected by this concern; the soft spot is specifically the 0p topology on which the thesis's own strongest remaining caveat rests.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The dissertation reports a search for an anomalous excess of single photons in MicroBooNE, targeting neutral-current Delta radiative decays (NC Delta to N gamma). It develops a joint Wire-Cell plus Pandora selection, evaluates backgrounds with GENIE simulation and a MicroBooNE tune, constrains systematic uncertainties with nu-mu-CC and NC-pi0 sidebands, and tests the MiniBooNE LEE under scaled NC Delta-like hypotheses characterized by parameters x_Delta and x_0p. The observed data are reported as consistent with the nominal prediction, and the analysis excludes parts of the (x_Delta, x_0p) plane while leaving the zero-visible-proton (0p) axis open, as stated in the abstract.","tokens_in":59337,"tokens_out":4559,"duration_ms":48667,"significance":"If the result holds, it disfavors photon-like explanations of the MiniBooNE LEE that would produce NC Delta radiative single photons accompanied by a visible proton, and it provides a detailed experimental template for single-photon searches in LArTPCs. The analysis has notable strengths: selection development on a small blinded subset (Sec. 3.3.3), nine detector systematic universes (Sec. 2.4), conditional constraints from nu-mu-CC and NC-pi0 sidebands (Sec. 4.5.3), fake-data closure tests (Appendix A), and a combination of two independent reconstruction chains. The external benchmark built from MiniBooNE's measured excess is not derived from MicroBooNE's own fit, so the exclusion test is not circular. The principal weakness is that the 0p topology, on which the thesis explicitly concedes no constraint, is also the place where the simulation-dependent prediction is least tested.","major_comments":[{"comment":"The central null statement 'data consistent with our nominal expectation' includes the 0p channel, but the true-0p efficiency and 0p shower efficiencies are presented as simulation-only quantities with no data closure test for that topology. The thesis itself documents a GENIE branching-ratio bugfix, material differences between GENIE FSI models for single photons (Figs. 4.39-4.43), and ad hoc Geant4 electromagnetic model modifications (Sec. 5.3.1). As written, the quoted consistency with expectation for the 0p component is therefore not robust: a genuine 0p photon excess could be absorbed into the simulated prediction if the true 0p efficiency or photonuclear absorption differs from the model. Please either add a data-driven 0p validation/sideband constraint or explicitly rescope the consistency claim to the visible-proton topology and quantify how large a 0p excess could be hidden within the quoted uncertainties.","section":"Sec. 4.5.1, Fig. 4.35; Secs. 5.2.2-5.2.3; Secs. 4.39-4.43; Sec. 5.3.1"},{"comment":"Section 4.6.5 introduces x_0p precisely because data cannot pin down the zero-proton fraction, and the resulting exclusion leaves the x_0p axis open. The conclusions should therefore always carry the qualifier 'with visible proton activity' when describing excluded NC Delta-like LEE hypotheses. The abstract already states this caveat, but the chapter-level summary and the captions of Figs. 4.73-4.75 should be equally explicit so that a reader does not walk away with the impression that 0p-only models are tested.","section":"Sec. 4.6.5, Figs. 4.73 and 4.75"},{"comment":"The conditional constraint uses nu-mu-CC and NC-pi0 sidebands. These channels share flux and cross-section systematics with the single-photon signal, but they do not directly constrain the 0p single-photon reconstruction efficiency or the photonuclear-absorption model for 0p photons. The post-constraint uncertainties in Table 4.4 may therefore be underestimated for the 0p channel. The text should state this limitation explicitly and either provide an unconstrained 0p systematic or demonstrate a quantitative correlation between the sideband channels and the 0p acceptance.","section":"Sec. 4.5.3, Table 4.4"}],"minor_comments":[{"comment":"Several typographical errors should be corrected, including 'instutitions', 'componentns', 'descripancy', and 'decomissioning'.","section":"Acknowledgments and throughout"},{"comment":"The caption reads 'tob, left, right, and bottom panels'; this should be 'top, left, right, and bottom panels'.","section":"Fig. 2.7 caption"},{"comment":"The phrase 'We finished finished processing' contains a duplicated word and should be revised.","section":"Sec. 2.4.4"},{"comment":"The notation for the signal process is inconsistent (NC Delta, NC Delta radiative decay, NC Delta to N gamma). Define the notation once and use it consistently throughout.","section":"Chapters 4-5"}],"recommendation":"major_revision","confidential_remarks":"The manuscript is a PhD thesis rather than a conventional journal article, and its methodological transparency is a real strength. My main concern is that the headline consistency claim is strongest where the simulation is best validated (visible-proton events) and weakest where the thesis admits no constraint (0p events). Because the abstract is honest about this, the issue is fixable by re-scoping and quantification rather than by new data. I would not reject on circularity grounds; the MiniBooNE LEE model is an external benchmark, and the sideband constraints are disclosed and standard."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Lee, here's the quick take. This arXiv posting is a University of Chicago PhD thesis, not a collaboration paper, and it reads like one: thorough to a fault, with the analysis details you'd want. The genuinely new pieces are the joint Wire-Cell plus Pandora selection for NC Delta radiative decay, the particle-multiplicity comparisons, and the 2D (x_delta, x_0p) exclusions against MiniBooNE LEE hypotheses. The rest of the thesis restates published MicroBooNE nu_e CC and sterile neutrino results, but that's fine for a thesis.\n\nWhat's done well: the methodology is visible and sound. Blinded BDT development on a small subset, nine detector systematic universes, conditional constraints on nu_mu CC and NC pi0 sidebands, and fake-data closure tests in the appendix. The thesis is also honest about its main caveat, which is the zero-proton (0p) topology. The abstract states it plainly: data are consistent with expectation, but 0p single-photon sources cannot be ruled out.\n\nThe soft spot is real but contained. The 'consistent with expectation' statement for 0p events rests on simulated 0p photon rates and reconstruction efficiencies. The thesis itself shows those models aren't settled — it documents a GENIE branching-ratio bugfix, spread among GENIE FSI models for photonuclear absorption, and ad hoc Geant4 EM model modifications. That doesn't by itself make the result wrong; it means the 0p channel can't carry much exclusion weight, and the thesis doesn't pretend it can. The visible-proton exclusion, which is the part with real discriminating power against Delta radiative decay explanations, is not affected by this concern. The x_0p parameter is a free parameter in the 2D exclusion for exactly this reason.\n\nOne practical caveat: the review copy I saw was truncated before section 4.6 and chapter 5, so I could not directly inspect the central hypothesis-test exclusions. What I saw supports the abstract; I'd want to see those sections before signing off. No code or data artifacts are shipped, which is normal for a thesis but limits reproducibility.\n\nWho should read this: anyone working on single-photon backgrounds in LArTPCs or on photon-like explanations of the MiniBooNE LEE would get value, especially from the efficiency studies and the honest treatment of 0p uncertainties.\n\nMy recommendation: send it to peer review. It deserves a serious referee, and the referee should focus on the 0p simulation systematics and ask for the truncated sections in full.","headline":"A solid, honest MicroBooNE doctoral thesis with a genuinely new single-photon selection and a null result whose visible-proton leg is robust and whose 0p leg is explicitly model-dependent.","tokens_in":60057,"tokens_out":2622,"would_cite":true,"duration_ms":27823,"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":"MicroBooNE finds no excess of single photons from neutrino-induced Delta radiative decays, but cannot rule out photon sources with no visible proton.","keywords":["single photon search","MicroBooNE","liquid argon time projection chamber","neutral current Delta radiative decay","MiniBooNE low energy excess","photon-like excess","neutrino anomaly","LEE hypothesis"],"falsifier":"A dedicated re-analysis of the same data that tags single-photon events with no reconstructed proton—for example by using a control sample of $e^+e^-$ pairs to measure the true-0p efficiency—and finds a significant low-energy excess would directly contradict the paper's claim of consistency with expectation. Concretely, recomputing the one-bin count with the true-0p efficiency doubled and photonuclear absorption reduced by 30% would show whether the nominal prediction is robust.","tokens_in":58714,"feed_emoji":"⚛️","tokens_out":7301,"duration_ms":69880,"temperature":0.7,"pith_summary":"MicroBooNE set out to test whether the anomalous low-energy excess seen by MiniBooNE could be made of photon showers rather than electron showers, and this thesis targets the largest expected source of single photons: neutral-current $\\Delta$ radiative decays. Using a joint Wire-Cell and Pandora selection, the analysis observes data consistent with nominal expectations and finds no significant excess of single photons. Hypothesis tests that scale the NC $\\Delta$ radiative decay rate to match MiniBooNE's LEE exclude large parts of the scaling plane when the photons come with a visible proton, but leave the no-visible-proton axis open. If the result holds, explanations of the MiniBooNE LEE that produce single photons with visible protons are disfavored at the tested energies; photon-like explanations with invisible protons remain viable.","feed_headline":"No photon excess in MicroBooNE's Delta decay search","feed_subtitle":"The result leaves the no-visible-proton photon explanation of MiniBooNE's anomaly open.","key_machinery":"The load-bearing object is the $NC\\Delta\\to N\\gamma$ single-photon selection, built by combining two independent LArTPC reconstructions: Wire-Cell 3D imaging and Pandora pattern recognition. The selection isolates events with one electromagnetic shower and uses electron-photon discrimination (a gap between vertex and shower, and the $dE/dx$ at the shower stem) to identify photons; the hypothesis tests then scan the scaling factor $x_\\Delta$ (multiplying the nominal NC $\\Delta$ radiative rate) and the fraction $x_{0p}$ of single photons with no visible proton, a parameter introduced because the data cannot constrain the true-0p efficiency. The machinery also includes the true-0p efficiency maps and the systematic covariance, which carry the analysis from reconstructed kinematics to the exclusion contours.","core_discovery":"The central claim is that the joint Wire-Cell + Pandora selection for neutral-current $\\Delta$ radiative decays ($NC\\Delta\\to N\\gamma$) in MicroBooNE finds data consistent with the nominal background prediction, with no significant excess of single-photon events. The analysis quantifies this by testing the 'enhanced NC $\\Delta$' version of the MiniBooNE LEE hypothesis, in which the NC $\\Delta$ radiative decay rate is scaled by a factor $x_\\Delta$, and a more general two-parameter hypothesis in which a fraction $x_{0p}$ of the photons have no visible proton. The data exclude large regions of the ($x_\\Delta$, $x_{0p}$) plane, particularly where the photons are accompanied by visible protons, but cannot exclude the $x_{0p}$-only axis. In the author's words, the data are consistent with expectation but cannot rule out all potential sources of additional single photons, especially those with no visible proton activity.","pith_inferences":["I read the 0p axis as the place where a photon-like LEE could still be hiding; a natural next step would be a selection that sacrifices visible-proton purity to recover low-energy 0p showers and uses the measured 0p efficiency as a fit parameter.","The same ($x_\\Delta$, $x_{0p}$) scaling-plane technique could be applied to other single-photon searches, such as coherent or inclusive single photons, to map out which photon production mechanisms are compatible with the null result.","Because GENIE's photonuclear absorption is a leading modeling uncertainty, a data-driven measurement of photon survival in argon would strengthen or overturn the conclusion; the thesis's own Geant4 EM modifications show this modeling is not settled.","If a future liquid-argon experiment with lower proton thresholds finds a 0p single-photon excess, it would revive the photon interpretation of the MiniBooNE LEE that this result leaves open."],"forward_implications":["If the central claim is right, the enhanced NC Delta radiative decay explanation of the MiniBooNE LEE is disfavored whenever the photon is accompanied by a visible proton at MicroBooNE's energies.","The two-parameter exclusion in ($x_\\Delta$, $x_{0p}$) means any NC Delta-like LEE model that requires a large visible-proton fraction is ruled out, while models living on the no-visible-proton axis survive.","The result demonstrates that joint Wire-Cell + Pandora single-photon selections can constrain photon-like LEE explanations and provides a reusable framework for such searches.","Data being consistent with nominal expectation implies no new single-photon source is needed to describe the current MicroBooNE sample, within the untested 0p region.","The 0p channel is the clear next target: the analysis's own efficiency maps show where future sensitivity must improve."],"supporting_citations":[{"why":"Supplies the MiniBooNE Low Energy Excess observation that motivates the search and provides the LEE hypothesis to test.","marker":"[73]"},{"why":"Supplies the Pandora reconstruction chain used for one of the two independent selections.","marker":"[160]"},{"why":"Supplies the Wire-Cell 3D reconstruction framework used for the other selection.","marker":"[161, 162]"},{"why":"Supplies the generic neutrino selection that rejects cosmic rays and defines the fiducial volume for the analysis.","marker":"[167]"},{"why":"Supplies the MiniBooNE LEE response matrix and unfolded prediction used to define the eLEE and LEE hypothesis models.","marker":"[174]"},{"why":"Supplies the GENIE v3.0.6 neutrino event generator used to simulate signal and background events.","marker":"[176]"},{"why":"Supplies the MicroBooNE tune for neutrino-argon cross-section modeling and its systematic variations.","marker":"[177]"},{"why":"Supplies the Geant4 Bertini intranuclear cascade model used for hadron re-interaction simulation and uncertainties.","marker":"[178]"}],"fun_headline_variants":["MicroBooNE sees no single-photon excess","No photon bump from Delta decays","MicroBooNE leaves photon mystery open for no-proton events","MicroBooNE data consistent with background in photon search"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The null result rests on the simulation's prediction for single photons that leave no visible proton—the topology the thesis explicitly says it cannot constrain—so if GENIE's photon production or photonuclear absorption, or the 0p reconstruction efficiency, is wrong, the apparent consistency could hide a real photon excess.","fun_headline_variants_meta":{"raw":{"variants":["MicroBooNE sees no single-photon excess","No photon bump from Delta decays","MicroBooNE leaves photon mystery open for no-proton events","MicroBooNE data consistent with background in photon search"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.001015,"raw_usage":{"total_tokens":4294,"prompt_tokens":959,"completion_tokens":3335,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":575,"completion_tokens_details":{"reasoning_tokens":3272}},"tokens_in":575,"tokens_out":3335,"duration_ms":27134,"temperature":1.0,"reasoning_tokens":3272,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-15T18:43:34.435716+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"A dedicated re-analysis of the same data that tags single-photon events with no reconstructed proton—for example by using a control sample of $e^+e^-$ pairs to measure the true-0p efficiency—and finds a significant low-energy excess would directly contradict the paper's claim of consistency with expectation. Concretely, recomputing the one-bin count with the true-0p efficiency doubled and photonuclear absorption reduced by 30% would show whether the nominal prediction is robust.","supporting_citations":[],"review_version":2}