{"id":"8a4b1e56-7f87-4e9a-be88-127455296014","arxiv_id":"2412.10150","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":1,"one_line_summary":"E137 beam-dump data exclude spin-2 dark matter mediator couplings from about 8e-8 to 1e-5 per GeV for mediator masses 100 to 800 MeV, assuming equal electron and photon couplings.","lead":"Physicists calculated how often a hypothetical heavy particle, a spin-2 dark matter mediator, would be produced when an electron beam hits a metal target, and used the old E137 experiment's null result to rule out a range of its couplings to electrons and photons. The work helps map where to look for such mediators in future fixed-target experiments and shows which approximation formulas are reliable.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Eq. (32) appears to give a G→γγ width a factor of 36 too small; using the correct width changes l_G in Eq. (33) by about a factor of 2.8, so the quoted E137 exclusion band needs re-evaluation.","rationale":"The reader's CONDITIONAL verdict is appropriate, and the reader's flagged assumption (E_e ≈ E_G in Eq. (33)) is a real concern about decay-length averaging. However, I find a more concrete and checkable issue in Eq. (32): the diphoton partial width appears to be a factor of 36 smaller than the standard result for a massive spin-2 mediator coupled to the photon energy-momentum tensor. Since the paper uses c_ee = c_γγ and the total width appears directly in the decay length that sets the visible-mode signal, this is a load-bearing algebraic point, not just a modeling assumption. The check is straightforward: re-derive the width from the provided Feynman rules or compare with the cited literature. If the width is indeed 36 times larger, the total decay rate is roughly 2.84 times larger than used, shortening the decay length and shifting both edges of the excluded coupling band. The qualitative exclusion of a band is unlikely to disappear, but the specific quantitative claim in the abstract would need revision. The reader's concern about energy averaging is complementary and should also be addressed, possibly with a Monte Carlo or a x-weighted decay-probability average. I therefore keep the CONDITIONAL verdict unchanged, pending the width verification and a quantitative estimate of the decay-energy averaging effect.","tokens_in":18503,"tokens_out":18369,"duration_ms":193829,"concrete_test":"Recompute Γ(G→γγ) using the Feynman rule of Eq. (7) (with m_V = 0) and the massive spin-2 polarization sum in Eq. (A4), or compare with the published result in Refs. [32,33]. If the correct width is c² m_G³/(80π), recompute the total width and the number of signal events in Eq. (33) for m_G = 100, 300, and 800 MeV and for c_ee = 8×10⁻⁸, 10⁻⁶, and 10⁻⁵ GeV⁻¹. Then re-plot the 90% C.L. exclusion boundary in Fig. 5 and check whether the quoted band 8×10⁻⁸–10⁻⁵ GeV⁻¹ remains valid; a shift by more than ~50% in c would require a corrected abstract.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central claim is the quantitative E137 exclusion range 8×10⁻⁸ GeV⁻¹ ≲ c_ee ≲ 10⁻⁵ GeV⁻¹. That range is obtained from Eq. (33), which uses the decay length l_G = (E_G/m_G)(1/Γ_tot) with Γ_tot = Γ_{G→ee} + Γ_{G→γγ}. Eq. (31) for Γ_{G→ee} matches the standard massive-spin-2 result, but Eq. (32) gives Γ_{G→γγ} = (1/3) c² m_G³/(960π) = c² m_G³/(2880π). The standard result from the same coupling structure (e.g., Refs. [32,33]) is Γ_{G→γγ} = c² m_G³/(80π), which is a factor of 36 larger. With the assumed c_ee = c_γγ, the true total width is about (1 + 2) × c² m³/(160π) = 3 c² m³/(160π), whereas the paper's value is (1 + 1/18) × c² m³/(160π) ≈ 1.056 c² m³/(160π). Thus l_G is overestimated by a factor ≈ 2.84. This shifts the decay-probability factor e^{-L_sh/l_G} − e^{-L_tot/l_G} in Eq. (33) and therefore changes both the lower edge (long-lived regime, where the probability is ∝ 1/l_G, and the bound scales as Γ_tot^{1/4}) and the upper edge (short-lived regime, where the bound is sensitive to l_G). The qualitative claim that E137 excludes a band may survive, but the specific numbers in the abstract and Fig. 5 would have to be revised.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper studies bremsstrahlung-like production of a massive spin-2 mediator in lepton fixed-target experiments, comparing the Weizsäcker-Williams (WW) approximation with an exact tree-level (ETL) calculation, and then uses the E137 null result to derive a 90% C.L. exclusion band for the mediator coupling to electrons and photons. The central quantitative claim is that E137 excludes 8×10^-8 GeV^-1 ≲ c_ee^G ≲ 10^-5 GeV^-1 for mediator masses 100 MeV ≲ m_G ≲ 800 MeV under the assumption c_ee^G = c_γγ^G. The paper also gives a brief comparison with the LDMX reach and with BaBar constraints.","tokens_in":1532,"tokens_out":1557,"duration_ms":100810,"significance":"If the central calculation is correct, the paper provides a new and useful exclusion for a simplified massive spin-2 mediator scenario, and the WW/ETL comparison for masses above roughly 200 MeV is a valuable cross-check for future fixed-target proposals. The E137 bound is an external null-result recast rather than a fit, so the circularity burden is low. The paper is also notable for giving explicit analytic expressions for the ETL amplitude squared and for the spin-2 vertices. However, the numerical E137 exclusion range is not robust until the photon decay width and the energy-spectrum treatment are corrected and quantified.","major_comments":[{"comment":"The decay width for G→γγ is given in Eq. (32) as Γ_{G→γγ} = (1/3) c² m_G³/(960π) = c² m_G³/(2880π). For the same Fierz-Pauli coupling to the photon energy-momentum tensor, the standard result is Γ_{G→γγ} = c² m_G³/(80π), a factor of 36 larger. This is not merely a convention issue: the numerical estimate for l_{G→γγ} stated in the text, 4.5×10^5 cm at E_G=10 GeV, m_G=100 MeV, and c=10^-6 GeV^-1, is consistent with the standard width and not with Eq. (32). Since l_G enters the decay-probability factor in Eq. (33) exponentially, the E137 exclusion band in Fig. 5 and the quoted range in the abstract must be re-evaluated; with c_ee=c_γγ the total width increases by a factor ≈54/19≈2.84, shifting the lower edge by about (54/19)^{1/4} and the upper edge by a comparable factor. The authors should correct Eq. (32) and recompute the limits.","section":"Eq. (32) and following paragraph"},{"comment":"The signal estimate in Eq. (33) assumes that the mediator carries essentially all of the beam energy, E_e ≈ E_G, when computing the decay probability, while the production cross section σ_tot is integrated over energy fractions x down to x_cut=0.1 for E137 (Table I). Because l_G is proportional to E_G, events with x≪1 have considerably shorter decay lengths, and the exponential factor in Eq. (33) is not correctly averaged over the production spectrum. The paper provides no quantification of the resulting shift in the exclusion band. The authors should either weight the decay probability over the ETL/WW differential cross section in x or justify explicitly that the spectrum is so sharply peaked near x≈1 that the approximation is accurate for the quoted bounds.","section":"Sec. V.B, Eq. (33)"},{"comment":"The ETL amplitude squared |A^G_{2→3}|² in Eq. (B1), with the coefficients in Eqs. (B3)–(B17), is central to both the WW/ETL comparison and the E137 limit, but it is presented without derivation and without an accompanying code or ancillary file. A reader cannot independently verify the expression, and the paper does not state which computer-algebra tool or method was used to obtain it. The authors should provide the derivation or a machine-readable ancillary file, and ideally a numerical cross-check against an independent evaluation at a representative phase-space point.","section":"Appendix B, Eq. (B1)"}],"minor_comments":[{"comment":"The abstract states the lower edge of the excluded coupling range as 8×10^-8 GeV^-1, while the conclusion states 10^-7 GeV^-1. This numerical discrepancy should be resolved, especially after the width correction is applied.","section":"Abstract vs. Conclusion"},{"comment":"The text refers to the small-mass regime as m_G ≲ 100 GeV; from context this should be 100 MeV.","section":"Sec. IV A"},{"comment":"There are several typographical and grammatical errors, e.g., 'has been ruled out the the couplings' in the abstract and 'the E137 experiment has been ruled out the the parameter space' in Sec. V C. These should be corrected in a revision.","section":"Throughout"},{"comment":"The production estimate N^{brem}_G uses a single target radiation length for E137, but the shielding and detector geometry are described only in words; it would be clearer to state explicitly that L_T^{E137}=X_0 and to define all lengths in one place.","section":"Eq. (29) and Sec. V B"}],"recommendation":"major_revision","confidential_remarks":"The main novelty of the paper is the E137 exclusion band, and the cross-section comparison, while useful, is incremental. The issues raised in the major comments are technical and load-bearing, but they appear fixable: the width in Eq. (32) is likely a typo given the paper's own decay-length estimate, and the x-averaging issue can be addressed by weighting the decay probability over the production spectrum. If the corrected numbers shift the quoted band but the qualitative exclusion survives, the result would be acceptable after revision. I do not see grounds for rejection, provided the authors supply the requested numerical checks and machine-readable amplitude."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Two things to know upfront. The WW/ETL comparison for spin-2 mediator production in lepton fixed-target experiments is a genuinely useful piece of work. The authors have done the exact tree-level calculation, checked it against the Weizsacker-Williams approximation across NA64e, NA64μ, LDMX, M3, and E137 parameters, and found agreement at the percent level for mG > 200 MeV. That is a real result for anyone planning dark-sector searches. The self-citations are to independent cross-section calculations, so no circularity issue.\n\nThe second thing is less friendly. The E137 visible-mode exclusion band in the abstract and Fig. 5 rests on Eq. (32), the partial width Γ(G→γγ). The paper quotes (1/3)c^2 m^3/(960π). The standard result from the same coupling c Gμν T^{μν} is c^2 m^3/(80π), i.e. 36 times larger. With c_ee = c_γγ, that means the total width in the paper is underestimated by about a factor of 2.8, and the decay length in Eq. (33) is overestimated by the same factor. The shape of the excluded band changes accordingly. The qualitative claim that E137 excludes a band may survive, but the specific numbers 8e-8 to 1e-5 GeV^-1 are not reliable as they stand.\n\nOther soft spots are secondary. Eq. (33) assumes Ee ≈ EG for the decay probability while the production cross section integrates down to xcut = 0.1; the error from that is not quantified. The amplitude squared in Appendix B is a long algebraic expression with no derivation or code, which makes the ETL calculation hard to verify. The low-mass edge of the band sits where WW and ETL already differ, so that boundary is less robust.\n\nThe paper is a competent recasting with one load-bearing mistake. It deserves a serious referee, but the referee should send it back for a corrected photon width, a re-evaluated E137 band, and ideally a quantitative estimate of the energy-loss effect. The WW/ETL comparison is worth keeping. I would not cite the E137 band in its current form.","headline":"Useful WW/ETL comparison for spin-2 mediators, but the E137 exclusion band is built on a photon width that looks 36x too small.","tokens_in":19405,"tokens_out":5713,"would_cite":false,"duration_ms":58354,"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":"The E137 beam-dump experiment rules out a massive spin-2 dark-matter mediator that couples to photons and electrons, for mediator masses between 100 MeV and 800 MeV and couplings in the range $8\\times10^{-8}$ to $10^{-5}$ GeV$^{-1}$.","keywords":["spin-2 mediator","dark matter mediator","bremsstrahlung-like production","fixed-target experiment","E137","Weizsacker-Williams approximation","exact tree-level","visible decay"],"falsifier":"Recompute the E137 visible-decay signal by weighting every produced mediator with its actual energy fraction $x=E_G/E_e$ drawn from the exact tree-level differential cross section $d\\sigma/dx$, and integrate the survival probability $e^{-L_{\\rm sh}/l_G(E_G)}-e^{-L_{\\rm tot}/l_G(E_G)}$ over $x$ from $0.1$ to $1$; if the resulting 90% confidence excluded band shifts noticeably from $8\\times10^{-8}\\,\\mathrm{GeV}^{-1}\\lesssim c_{ee}^G\\lesssim10^{-5}\\,\\mathrm{GeV}^{-1}$, the central claim as stated would not hold at that precision.","tokens_in":18328,"feed_emoji":"⚛️","tokens_out":4121,"duration_ms":44642,"temperature":0.7,"pith_summary":"The paper studies a massive spin-2 particle as a mediator between Standard Model particles and dark matter, produced in lepton fixed-target collisions through a bremsstrahlung-like process. It compares two calculational schemes, the Weizsacker-Williams approximation and the exact tree-level approach, for several experiments and finds agreement at the level of a few percent for mediator masses above roughly 200 MeV. Its central new result is an exclusion from the E137 electron beam-dump experiment: with $1.87\\times10^{20}$ electrons on target and zero observed signal events, the paper claims the couplings of a spin-2 mediator that couples universally to electrons and photons are ruled out in the band $8\\times10^{-8}\\lesssim c_{ee}^{G}\\lesssim10^{-5}$ GeV$^{-1}$ for mediator masses $100\\,\\mathrm{MeV}\\lesssim m_G\\lesssim800\\,\\mathrm{MeV}$. A sympathetic reader would care because this turns an old null result into a concrete constraint on a relatively unexplored spin-2 portal scenario in the sub-GeV mass range.","feed_headline":"E137 rules out a spin-2 dark-matter mediator","feed_subtitle":"A 20 GeV electron beam dump finds no photon or electron decays, excluding couplings $8\\times10^{-8}$ to $10^{-5}$ GeV$^{-1}$ for 100–800…","key_machinery":"The central object is the massive spin-2 mediator field $G_{\\mu\\nu}$ coupled to Standard Model fields through their energy-momentum tensors, with coupling constants $c_i^G$ of dimension GeV$^{-1}$. The calculation that carries the argument is the bremsstrahlung-like production process $lN\\to lNG$, evaluated either by the exact tree-level matrix element or by the Weizsacker-Williams approximation that reduces it to a Compton-like $l\\gamma^*\\to lG$ subprocess with a virtual photon flux. The signal estimate then uses the decay lengths of the mediator in the lab frame, $l_G = (E_G/m_G)(1/\\Gamma_{\\rm tot}^G)$, together with the thick-target formula, to count how many produced mediators survive through the shielding and decay inside the E137 fiducial volume; the Tsai-Schiff nuclear form factor is used throughout for the target.","core_discovery":"The paper establishes that the E137 null result excludes the simplified massive spin-2 mediator scenario with universal couplings to electrons and photons, $c_{ee}^G = c_{\\gamma\\gamma}^G$, over a specific coupling-mass band. The exclusion is derived by computing the number of visible decays $G\\to\\gamma\\gamma$ and $G\\to e^+e^-$ expected from bremsstrahlung-like production of the mediator, using both the exact tree-level cross section and the Weizsacker-Williams approximation, and comparing with the observed zero events under a 90% confidence level Poisson assumption. As a supporting technical claim, the paper shows that the Weizsacker-Williams and exact tree-level total cross sections agree at the $O(1)$ percent level for mediator masses $m_G\\gtrsim200$ MeV across the fixed-target experiments considered, while discrepancies above 50% appear for light mediators below about 100 MeV because the exact amplitude contains terms that grow as $1/m_G^2$ and $1/m_G^4$.","pith_inferences":["Because the decay-length averaging in the signal formula assumes the mediator carries essentially all of the beam energy ($E_e\\simeq E_G$) while the production cross section is integrated down to $x_{\\rm cut}=0.1$, the lower boundary of the excluded coupling band could shift once the actual energy distribution is folded in; a re-analysis with energy-weighted decay probabilities would sharpen the b","The same exact tree-level machinery can be applied to the muon-beam experiments NA64$\\mu$ and M3 in visible mode, which would produce analogous exclusions on a muonphilic spin-2 mediator coupling $c_{\\mu\\mu}^G$.","The good Weizsacker-Williams and exact tree-level agreement for heavy mediators means future high-statistics fixed-target searches can use the cheaper Weizsacker-Williams cross section in Monte Carlo generators without introducing percent-level bias.","Combining the E137 exclusion with relic-density curves for vector dark matter leaves a narrower surviving parameter window, and the paper's result removes one of the few remaining sub-GeV spin-2 thermal targets."],"forward_implications":["The E137 experiment excludes spin-2 mediator couplings $8\\times10^{-8}\\lesssim c_{ee}^G\\lesssim10^{-5}$ GeV$^{-1}$ for masses $100\\,\\mathrm{MeV}\\lesssim m_G\\lesssim800\\,\\mathrm{MeV}$, assuming universal couplings to electrons and photons and visible decays.","The Weizsacker-Williams approximation is reliable at the percent level for mediator masses above roughly 200 MeV, so it can be used for future sensitivity projections in this mass range.","For mediator masses below about 100 MeV the Weizsacker-Williams approximation disagrees with the exact tree-level result by more than 50% and should not be trusted there.","The projected visible-mode sensitivity of LDMX with $10^{15}$ electrons on target is already covered by the BaBar mono-photon constraint for $m_G\\lesssim7$ GeV and $c_{ee}^G\\lesssim3\\times10^{-5}$ GeV.","The E137 exclusion also rules out a vector dark-matter benchmark with $m_V\\simeq300$ MeV for couplings around $10^{-7}\\lesssim c_{ee}^G\\lesssim3\\times10^{-6}$ GeV$^{-1}$. "],"supporting_citations":[{"why":"Supplies the E137 experiment setup, shielding lengths, and null result that the central exclusion is built on.","marker":"[83]"},{"why":"Provides the exact tree-level bremsstrahlung cross-section formalism and the E137 analysis method that this paper extends to the spin-2 mediator.","marker":"[46]"},{"why":"Provides the thick-target treatment and E137 shielding parameters used to convert production into visible decay events.","marker":"[47]"},{"why":"Gives the thick-target approximation formula for the number of signal events in visible mode that the exclusion count uses.","marker":"[85]"},{"why":"Supplies the partial decay widths of the spin-2 mediator to $e^+e^-$ and $\\gamma\\gamma$, which set the decay lengths in the signal estimate.","marker":"[32]"},{"why":"Defines the benchmark spin-2 mediator model with universal couplings and provides the thermal relic-density curves used in the comparison.","marker":"[33]"},{"why":"Provides the Weizsacker-Williams approximation and the Tsai-Schiff form factor used for the photon flux from the nucleus.","marker":"[75]"},{"why":"Supplies the preceding spin-2 mediator cross-section calculations and amplitude squared that the exact tree-level evaluation builds on.","marker":"[44]"}],"fun_headline_variants":["E137 excludes spin-2 mediator couplings","Spin-2 mediator ruled out by E137 beam dump","E137 closes window on spin-2 dark mediator","No spin-2 mediator signal in E137 data","E137 tightens spin-2 mediator constraints"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The signal estimate assumes that the produced mediator carries essentially all of the beam energy when computing its decay length, even though the production cross section is integrated over mediator energy fractions down to $x_{\\rm cut}=0.1$; if a significant fraction of events have much lower mediator energy, the decay-length averaging changes and the excluded coupling band shifts.","fun_headline_variants_meta":{"raw":{"variants":["E137 excludes spin-2 mediator couplings","Spin-2 mediator ruled out by E137 beam dump","E137 closes window on spin-2 dark mediator","No spin-2 mediator signal in E137 data","E137 tightens spin-2 mediator constraints"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.00028,"raw_usage":{"total_tokens":1732,"prompt_tokens":1086,"completion_tokens":646,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":702,"completion_tokens_details":{"reasoning_tokens":572}},"tokens_in":702,"tokens_out":646,"duration_ms":6555,"temperature":1.0,"reasoning_tokens":572,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-11T16:18:35.571322+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Recompute the E137 visible-decay signal by weighting every produced mediator with its actual energy fraction $x=E_G/E_e$ drawn from the exact tree-level differential cross section $d\\sigma/dx$, and integrate the survival probability $e^{-L_{\\rm sh}/l_G(E_G)}-e^{-L_{\\rm tot}/l_G(E_G)}$ over $x$ from $0.1$ to $1$; if the resulting 90% confidence excluded band shifts noticeably from $8\\times10^{-8}\\,\\mathrm{GeV}^{-1}\\lesssim c_{ee}^G\\lesssim10^{-5}\\,\\mathrm{GeV}^{-1}$, the central claim as stated would not hold at that precision.","supporting_citations":[{"cited_title":"Tsai, Rev","cited_arxiv_id":null,"evidence_quote":"Provides the Weizsacker-Williams approximation and the Tsai-Schiff form factor used for the photon flux from the nucleus."}],"review_version":1}