{"id":"14acaad5-8db4-45de-8d94-899b4a0fe25a","arxiv_id":"2505.00124","paper_version":1,"verdict":"CONDITIONAL","confidence":"HIGH","novelty_score":1.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":0,"one_line_summary":"Review and tutorial on collider searches for axions and ALPs, with a simplified reproduction of the Belle II photophilic ALP search.","lead":"This paper is a set of lecture notes that introduces axion and axion-like particle (ALP) searches at colliders, ending with a worked tutorial that reproduces the Belle II search for photophilic ALPs. It is useful as a pedagogical reference, but it contains no new scientific result.","discovery_kind":"review","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Eq. (54) undercounts Γ(a→γγ) by a factor of 2 relative to Eq. (25) and the standard formula, and this error propagates into the lifetimes and the Fig. 19 reach curves, weakening the tutorial's claim to reproduce Belle II.","rationale":"I read the paper in good faith as a pedagogical review and tutorial whose central claim is that the worked example reproduces the published Belle II analysis of Ref. [82]. The tutorial's structure is sound: it gives the Lagrangian, derives the production cross section, identifies four signal categories, and walks through the detector parameters and reach calculation. The simplified detector model and background treatment are explicitly stated as approximations, so I do not treat them as the primary weakness. The factor-of-two error in Eq. (54) is more load-bearing because it is an internal inconsistency: Eq. (25) is the standard width in the same normalization, and the two equations cannot both be correct. The error is concrete and propagates into the lifetimes, decay lengths, survival probabilities, and ultimately the Fig. 19 reach curves, so it directly affects the claim that the tutorial reproduces the Belle II sensitivity. The reader's rationale already flagged this error, but the reader's stated weakest_assumption was the detector/background idealization; hence partial agreement. The paper's independent value as a tutorial is not destroyed by a factor-of-two error that can be corrected locally, so I do not see grounds to move beyond the reader's CONDITIONAL verdict. The proposed concrete test is deliberately narrow: recompute the width analytically and regenerate the reach curves with the corrected lifetime to confirm the quantitative impact.","tokens_in":38310,"tokens_out":21159,"duration_ms":211956,"concrete_test":"Independently recompute Γ(a→γγ) from the Feynman rule in Eq. (45), using the standard two-body phase-space integral including the 1/2! Bose symmetry factor and the 1/(2m_a) prefactor, and verify whether the result is g_{aγγ}^2 m_a^3/(64π) or g_{aγγ}^2 m_a^3/(128π). Then propagate the corrected width through Eqs. (56)–(58) and Eq. (72), regenerate the Fig. 19 reach curves for both L = 445 pb^-1 and L = 50 ab^-1, and compare the displaced and invisible boundaries with the published Belle II limits in Ref. [82]. If the corrected curves differ by more than about 10% in g_{aγγ}, the tutorial's quantitative reproduction claim requires revision.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The most load-bearing concern is an internal inconsistency in the tutorial's central calculation. Under the mapping between Eq. (14) and Eq. (44), g_{aγγ} = α C_γγ^{eff}/(2π f_a), the standard width Γ(a→γγ) = g_{aγγ}^2 m_a^3/(64π) reproduces Eq. (25) exactly. Eq. (54) instead gives g_{aγγ}^2 m_a^3/(128π), a factor of 2 too small. The missing factor is the identical-particle phase-space factor (1/2!) and/or the 1/(2m_a) prefactor in the two-body decay rate, which the heuristic 4π/(32π^2) prefactor in Eq. (54) does not fully capture. Because τ_a = Γ_a^{-1} enters Eqs. (56)–(58), the survival probability in Eq. (72), and therefore the invisible-signal reach in Fig. 19, this is not a harmless typo: it shifts the predicted event rates and the derived bounds. The inconsistency is internal, since Eq. (25) and Eq. (54) are supposed to describe the same width in the same normalization, so it does not depend on any assumption about detector modeling or backgrounds. The detector simplifications in §5.6–§5.8 are explicitly declared and are acceptable for a pedagogical tutorial, but the decay-width error is not flagged and propagates directly into the quantitative claims.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"These lecture notes provide an introduction to axion and ALP searches at colliders. After reviewing basic collider physics and the landscape of dark-sector portals, the authors focus on the QCD axion and ALP phenomenology, including production and decay, and give a broad overview of search strategies. The final part is a worked tutorial that starts from the Lagrangian in Eq. (44), derives the associated-production cross section, the ALP decay width, and the expected detector signatures, and ends with Fig. 19, which shows approximate Belle II exclusion bounds for both the invisible and resolved γγ signatures. The tutorial is explicitly based on the published Belle II analysis of Ref. [82].","tokens_in":38602,"tokens_out":13224,"duration_ms":137528,"significance":"The paper fills a useful pedagogical niche: it takes the reader from first principles to a concrete, reproducible estimate of the Belle II sensitivity to photophilic ALPs, benchmarking each step against a real analysis. The main strengths are the clear step-by-step derivations (including the 2→2 phase-space counting, the χ2 statistics detour, and the explicit detector-geometry description), the explicit statement of the simplifications, and the transparent comparison with published data points in Fig. 21. If the numerical error identified below is corrected, the tutorial will be a valuable resource for students entering the field. The paper does not claim to derive new bounds; its value is didactic, and the declared approximations (zero background for the invisible channel, perfect background subtraction for the resolved channel, 3-event discovery rule) are acceptable for that purpose provided the underlying widths are correct.","major_comments":[{"comment":"Eq. (54) gives Γ(a→γγ) = g²_{aγγ} m³_a/(128π), but the standard result with the convention in Eq. (44) is Γ(a→γγ) = g²_{aγγ} m³_a/(64π). The inconsistency is internal: under the mapping g_{aγγ} = α C^{eff}_{γγ}/(2π f_a) in §4.3, Eq. (25) reproduces exactly the 64π result, while Eq. (54) is a factor of 2 smaller. The missing factor is the identical-particle phase-space factor 1/2!, which is not fully captured by the heuristic prefactor (4π/32π²) in Eq. (54). This error propagates directly into Eq. (56) for the proper lifetime, Eq. (57) for the boosted decay length, the numerical example in Eq. (58), and—through the survival probability P(L) in Eq. (72)—into the reach curves in Fig. 19. Since the tutorial’s stated purpose is to reproduce the quantitative sensitivity of Belle II, this factor must be corrected throughout §5.3 and §5.7–§5.8, and the affected numerical results and Fig. 19 should be updated.","section":"§5.3, Eq. (54)"}],"minor_comments":[{"comment":"The word “desribed” in the phase-space paragraph is a typo and should be “described”.","section":"§5.2, after Eq. (52)"},{"comment":"The explanation of the prefactor in Eq. (54) would be clearer if the identical-particle factor 1/2! were mentioned explicitly; this is directly related to the factor-of-2 error and should be part of the corrected derivation.","section":"§5.3, Eq. (54)"},{"comment":"The sentence “the direction of the other photon in the rest frame is −(π−θ′)” is confusing; a photon back-to-back with a photon at polar angle θ′ has polar angle π−θ′ and an azimuthal angle shifted by π, so the notation with a minus sign should be revised.","section":"§5.4, Eq. (60)"},{"comment":"“The ECL barell” is a typo; it should be “the ECL barrel”.","section":"§5.6"}],"recommendation":"major_revision","confidential_remarks":"The factor-of-two error in Eq. (54) is the only substantive technical issue I found. It is a simple oversight, but because the tutorial explicitly claims to reflect the Belle II analysis and the error feeds directly into the shown reach curves, it must be corrected before publication. The self-citations to Zupan are used only as background references and do not affect the novel tutorial content. The paper is otherwise a solid and clearly written set of lecture notes; with the width fixed, it would be a valuable contribution to the PoS proceedings."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"This is a well-written set of lecture notes, not a new research result. Its value is pedagogical: a broad review of dark-sector searches at colliders plus a worked tutorial that walks a student from the ALP-photon Lagrangian to Belle II reach curves. The tutorial mostly works, but it contains a real internal inconsistency: Eq. (54) gives Gamma(a->γγ) = g^2 m^3/(128π), while the paper's own Eq. (25) and the standard result require g^2 m^3/(64π). That factor of two propagates into the lifetimes and the reach curves, so it is not a harmless typo. A student following the steps will get the wrong decay length by a factor of two, and the Fig. 19 limits inherit the shift.\n\nThe review part is up-to-date and clearly organized. The sections on kaon decays, beam dumps, and the flavor anarchy vs MFV comparison are instructive. The tutorial is a genuine step-by-step, including MadGraph usage, a chi-square detour, and explicitly declared detector simplifications. The citation pattern looks reasonable; self-citations are background references, not self-promotion.\n\nThe main soft spot is the decay-width error. There are also smaller issues: the simplifications in Sec. 5.6–5.8 (zero background, 3-event rule, perfect background subtraction) are fine for teaching, but the resulting reach should be labeled as order-of-magnitude, and the authors could say that more prominently. The degree-of-freedom counting in Eq. (17) is cryptic, and the discussion of ECL angular coverage could be clearer. These are minor.\n\nWho is this for? Graduate students in particle physics, especially those starting in axion/ALP phenomenology. It does not claim a new result and should not be judged as one. With the width formula corrected, it would be a dependable tutorial. I would send it to peer review—a referee's job here is to catch exactly this kind of error—and after a minor revision it is publishable as lecture notes or proceedings. I would not cite it in a research paper, but I would suggest it to students.","headline":"Useful lecture notes for students, but the tutorial contains a genuine factor-of-two error in the a->γγ width that should be corrected before publication.","tokens_in":39073,"tokens_out":2316,"would_cite":false,"duration_ms":24562,"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":"A photophilic axion, from Lagrangian to Belle II exclusion plot.","keywords":["axion-like particles","collider searches","photophilic ALP","Belle II","missing energy signatures","displaced vertices","dark sector portals","lecture notes"],"falsifier":"Compare the tutorial's Fig. 19 curves with the full Belle II exclusion contours from Ref. [82] at the same luminosity: if the resolved-channel contour disagrees by more than the expected reconstruction-efficiency factor in any mass bin, or if the invisible-channel contour excludes points the experiment does not, the simplified background treatment is the reason.","tokens_in":38106,"feed_emoji":"⚛️","tokens_out":8193,"duration_ms":83880,"temperature":0.7,"pith_summary":"These lecture notes argue that a collider search for axion-like particles can be carried end to end, from a two-line Lagrangian to an exclusion plot, and they demonstrate it for a photophilic ALP at an electron-positron collider. The paper walks through the production mechanisms, the decay width and boosted lifetime, and the four signal categories that emerge depending on where and how the ALP decays. It then reproduces the main ingredients of the Belle II search with 445 inverse picobarns of data, deriving reach curves for the invisible and resolved diphoton signatures with a simplified detector model. If the tutorial's recipe is right, a student or phenomenologist can estimate the sensitivity of a given e+e- experiment to a photophilic ALP without a full experimental simulation, and can see which signatures cover which regions of mass-coupling space.","feed_headline":"From Lagrangian to exclusion plot: Belle II's axion reach","feed_subtitle":"A worked tutorial derives visible, displaced, merged, and invisible signals, then reproduces the early-data search from scratch.","key_machinery":"The load-bearing object is the simplified Belle II detector model combined with the analytic ALP kinematics. The vertex resolution $L_{\\min}=0.14$ m and detector outer radius $L_{\\max}=1.55$ m are compared with the boosted decay length $\\ell_a = (|\\vec p_a|/m_a)\\,128\\pi/(g_{a\\gamma\\gamma}^2 m_a^3)$; the electromagnetic calorimeter's $0.8^\\circ$ angular resolution is compared with the opening angle $\\Delta\\theta \\sim 4m_a/\\sqrt{s}$. This comparison alone assigns each point in the mass-coupling plane to one of four signal classes. The statistical treatment is a $\\chi^2$ test where resolved-signal limits use a 90% CL with $\\Delta\\chi^2 = 2.71$, and invisible-signal limits use the $N_{\\rm signal}=3$ rule.","core_discovery":"The central claim is that the expected signals of a photophilic ALP at an e+e- collider are completely determined, in their parametric behavior, by two formulas and one geometric comparison. The cross section for $e^+e^- \\to \\gamma a$ scales as $\\frac{\\alpha g_{a\\gamma\\gamma}^2}{32}(1+\\cos^2\\theta)(1 - m_a^2/s)^3$, and the decay width is $\\Gamma(a\\to\\gamma\\gamma)=g_{a\\gamma\\gamma}^2 m_a^3/(128\\pi)$, so the boosted decay length is $\\ell_a \\sim (|\\vec p_a|/m_a)\\,128\\pi/(g_{a\\gamma\\gamma}^2 m_a^3)$. Comparing $\\ell_a$ to the detector's inner and outer radii, and comparing the typical photon opening angle, $\\sim 4m_a/\\sqrt{s}$, to the angular resolution, partitions the $(m_a, g_{a\\gamma\\gamma})$ plane into invisible, merged, displaced-resolved, and prompt-resolved regions. Using the Belle II geometry ($L_{\\min}=0.14$ m, $L_{\\max}=1.55$ m, angular resolution $0.8^\\circ$, $0.25$ GeV energy threshold), an idealized background treatment, and a three-event discovery threshold, the paper produces approximate 90% CL exclusion curves for 445 inverse picobarns and for the full 50 inverse attobarns, mirroring the published Belle II analysis.","pith_inferences":["The same closed-form chain can be rerun for any other electron-positron machine by changing $\\sqrt{s}$, $L_{\\min}$, $L_{\\max}$, and angular resolution, so the tutorial effectively gives a parametric calculator for photophilic-ALP reach rather than a single experiment-specific result.","The merged-photon region is classified but not quantified in the tutorial; a dedicated search using electromagnetic-shower shapes in that corner of parameter space would be a natural complement to the resolved and invisible analyses.","For a hadrophilic or leptophilic ALP, the same steps apply with the lifetime formulas of Section 4.3 replacing the two-photon width; the reach curves would be shaped by whichever decay channel dominates.","The idealized background treatment makes a concrete prediction: the full Belle II limits, which use a complete simulation, should lie close enough to the tutorial's curves to confirm that the simplified model captures the main sensitivity."],"forward_implications":["The resolved diphoton signature is the relevant probe for ALP masses from a few hundred MeV up to near $\\sqrt{s}$, where the production cross section vanishes as $(1 - m_a^2/s)^3$.","The invisible single-photon signature covers low ALP masses and small couplings, where the decay length exceeds the detector size; its reach improves only with integrated luminosity, roughly as an order of magnitude when going from 445 inverse picobarns to 50 inverse attobarns.","Photon-fusion production has a larger cross section than associated production at Belle II, but it is discarded because the outgoing electrons remain in the uninstrumented forward region and the resulting signature suffers large beam-induced backgrounds.","Whether a given ALP is visible, displaced, merged, or invisible is fixed by the ratios $\\ell_a/L_{\\rm det}$ and $\\Delta\\theta/\\Delta\\theta_{\\rm res}$, so the same classification applies to any e+e- detector with different radii and angular resolution.","For two benchmark flavor structures, rare kaon decays probe axion scales that differ by orders of magnitude: flavor-anarchic couplings reach $f_a \\sim 10^{12}$ GeV, while loop-generated minimal-flavor-violation couplings reach only $f_a \\sim 10^6$-$10^7$ GeV, or about 10 TeV for a gluon-only UV coupling."],"supporting_citations":[{"why":"The published Belle II search for ALPs in e+e- collisions that the tutorial's detector model, background assumptions, and reach curves are designed to reproduce.","marker":"[82]"},{"why":"Supplies the classification of visible, displaced, merged, and invisible ALP signals and the sensitivity estimates from which Fig. 18 is adapted.","marker":"[85]"},{"why":"The simulation tool used in the tutorial to compute photon-fusion production and the QED three-photon background cross sections numerically.","marker":"[83]"},{"why":"The statistics review providing the chi-square and confidence-level formalism used to set the 90% CL bounds.","marker":"[84]"},{"why":"Provides the low-energy effective theory for axions and ALPs, including the loop functions that enter the ALP-photon coupling in Eq. (26).","marker":"[72]"},{"why":"Source for the rare-kaon-decay reach estimates and the MFV loop formulas that set the $f_a$ scales quoted in Section 4.3.1.","marker":"[74]"},{"why":"Describes the Belle II detector whose ECL and SVD geometry define $L_{\\min}$, $L_{\\max}$, angular resolution, and energy threshold in the tutorial.","marker":"[13]"},{"why":"Gives the QCD axion mass relation and the chiral perturbation theory contribution to the ALP-photon coupling that fix the axion/ALP distinction.","marker":"[6]"}],"fun_headline_variants":["Two formulas map every axion signal at Belle II","Axion search tutorial ends at Belle II exclusion plot","Photophilic ALP signals from a decay length and opening angle","Distilling Belle II axion search into one geometric comparison"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The reach curves stand or fall on the assumption that a simplified detector model—inner radius 0.14 m, outer radius 1.55 m, 0.8 degree angular resolution, 0.25 GeV photon threshold—together with an idealized background treatment (zero background for the invisible channel, perfect background subtraction for the resolved channel, and a three-event discovery rule) reproduces the actual Belle II sensitivity.","fun_headline_variants_meta":{"raw":{"variants":["Two formulas map every axion signal at Belle II","Axion search tutorial ends at Belle II exclusion plot","Photophilic ALP signals from a decay length and opening angle","Distilling Belle II axion search into one geometric comparison"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.00018,"raw_usage":{"total_tokens":1307,"prompt_tokens":954,"completion_tokens":353,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":570,"completion_tokens_details":{"reasoning_tokens":287}},"tokens_in":570,"tokens_out":353,"duration_ms":4055,"temperature":1.0,"reasoning_tokens":287,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-16T04:50:38.445990+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Compare the tutorial's Fig. 19 curves with the full Belle II exclusion contours from Ref. [82] at the same luminosity: if the resolved-channel contour disagrees by more than the expected reconstruction-efficiency factor in any mass bin, or if the invisible-channel contour excludes points the experiment does not, the simplified background treatment is the reason.","supporting_citations":[{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"The statistics review providing the chi-square and confidence-level formalism used to set the 90% CL bounds."}],"review_version":1}