{"id":"3c19a731-66da-4415-afb7-d2b39a486d56","arxiv_id":"2411.15686","paper_version":1,"verdict":"REJECT","confidence":"HIGH","novelty_score":4.0,"correctness_risk":"high","formal_verification":"none","parameter_count":9,"one_line_summary":"Fits of a contact-interaction parameter to LEP Bhabha data yield deviations above 3σ in several models, but the overall fits are poor and only statistical uncertainties are considered.","lead":"Using published LEP electron-positron scattering data, the authors test whether the electron has a finite size by fitting a contact-interaction strength parameter to measured cross sections. They report some deviations above three standard deviations in differential cross sections, which they say could hint at new physics, though the analysis uses only statistical errors and has very poor overall fit quality.","discovery_kind":"replication","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The claimed >3σ signal is computed with statistical uncertainties only, despite Eq. (13) including systematics; this is the load-bearing weakness.","rationale":"The reader's weakest assumption—that the significance calculation relies on statistical uncertainties only—is exactly the most load-bearing issue. The paper's own Eq. (13) calls for adding systematic uncertainties in quadrature, yet Section 5 explicitly says the results include only statistical uncertainties. Since the entire new-physics claim rests on deviations of a few σ, omitting systematic uncertainties of the size typical for LEP differential cross-section measurements can easily reduce the significance below the 3σ threshold. This is not a matter of interpretation or model choice; it is a direct mismatch between the stated method and the executed analysis. The poor fit quality (reader-reported χ²/ndf ≈ 4.7) reinforces the concern because a large χ² indicates the fitted model does not adequately describe the data, making the extracted ε less interpretable as a signal. The lack of any reported χ²/ndf or significance in Table 1 further prevents the reader from independently assessing the claim. Because this concern is sufficient to invalidate the central claim, the reader's REJECT verdict is appropriate and no verdict adjustment is needed.","tokens_in":4942,"tokens_out":2351,"duration_ms":24461,"concrete_test":"Recompute the χ² and best-fit ε for the LR/RL and VV/AA models using the published statistical and systematic uncertainties (including bin-to-bin correlations where available) for the OPAL 189–209 GeV differential bins and the DELPHI/L3 bins. If the significance of the best-fit ε drops below 3σ, or if ε becomes consistent with zero at 1σ, the central claim fails. If the >3σ significance survives the inclusion of systematics, the concern is resolved.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The paper's central claim—that differential Bhabha data favor a nonzero contact interaction—stands or falls on the size of the quoted deviations. Section 4 defines χ² with both σ_Stat and σ_Sys in Eq. (13), but Section 5 states: 'For now the results only includes statistical uncertainty.' This is an internal inconsistency, and it is decisive because the abstract's phrase '(statistical uncertainty only)' concedes that the reported deviations are not robust to systematic effects. LEP differential cross-section measurements (OPAL, DELPHI, L3) carry correlated systematic uncertainties from luminosity, acceptance, and radiative corrections that are typically comparable to, or larger than, the statistical uncertainties in several bins. Neglecting them inflates the significance of any ε shift. Moreover, the table reports fitted ε and Λ values but no χ²/ndf or p-value for the nine models, so the abstract's 'some derivation over 3σ' cannot be checked from the paper alone. The reader's independently noted χ²/ndf ≈ 4.7 shows the overall fit is poor; a model that does not describe the data cannot provide a reliable significance for its parameter. Without systematics, correlations, or a look-elsewhere penalty across the nine model choices, the reported >3σ deviations do not establish that a contact interaction exists.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"This paper fits a single contact-interaction parameter ε (conventionally Λ = 1/√|ε|, via Eq. (12)) to published LEP Bhabha-scattering data (OPAL at 189-207 GeV plus DELPHI and L3 bins, 125 differential bins in total) by minimizing the χ² defined in Eq. (13), with the Standard Model prediction computed using the Babayaga@NLO generator. For total cross sections the fit is reported as compatible with ε = 0. For differential cross sections the abstract claims deviations over 3σ (statistical uncertainties only), interpreted as a hint that a contact interaction might exist; Table 1 lists best-fit ε values (0.0003-0.0055 TeV⁻²) and Λ values (13-31 TeV) for nine helicity models, and Figures 1-9 show the corresponding χ² parabolas with the fitted significances printed in the figure headers. The central claim, evaluated in the paper's own statistical framework, is that differential LEP Bhabha data prefer a nonzero ε.","tokens_in":5203,"tokens_out":19549,"duration_ms":169178,"significance":"If the claimed deviations were statistically sound, the paper would be reporting a discovery-scale preference for new physics at scales of 13-31 TeV in the purely leptonic process e+e- → e+e-, a falsifiable prediction relevant for future e+e- colliders, and a result in direct tension with the SM-consistent conclusions of the original OPAL, DELPHI, and L3 publications. It is a strength of the paper that it uses public data, a standard contact-interaction formalism, a modern MC generator (Babayaga@NLO), and transparent one-dimensional fits whose parabolas are printed in full; the authors also disclose in Section 5 that only statistical uncertainties are included. These strengths do not salvage the claim, however: the quoted significances are computed against a model that is itself rejected by the data at extremely large χ², the analysis drops the systematic uncertainties that Eq. (13) promises to include, no look-elsewhere penalty is applied for the nine models, and the numerical results are not reproducible from the text. As it stands, the paper does not provide evidence for a contact interaction.","major_comments":[{"comment":"The analysis is internally inconsistent on the treatment of uncertainties. Eq. (13) defines χ² with the denominator σ²_Stat + σ²_Sys, but Section 5 states 'For now the results only includes statistical uncertainty'; all of Table 1 and the significance claims in the abstract are computed with the statistical-only version. The published OPAL, DELPHI, and L3 differential cross sections carry correlated systematic uncertainties (luminosity normalization, selection efficiencies, radiative-correction uncertainties) that in several bins are comparable to or larger than the statistical errors, so dropping σ_Sys systematically inflates the significance of ε. Because the abstract's conclusion that a contact interaction 'might exist' rests entirely on this inflated significance, the omission is load-bearing for the paper's central claim and must be repaired with the full published systematic uncertainties, including bin-to-bin correlations.","section":"Section 5 vs Eq. (13)"},{"comment":"The absolute quality of the fits is never reported, which is decisive for the interpretation. The parabolas in Figures 1-9 have minima c between 523.7 and 587.1 for 125 differential bins (ndf = 124 after fitting one parameter), i.e., χ²/ndf ≈ 4.2-4.7 for every model. Even at the best-fit ε, the contact-interaction model is inconsistent with the quoted errors (a χ² of 524-587 with 124 dof is more than 25 standard deviations above the mean of the χ² distribution), so the improvement over ε = 0 (Δχ² of about 0.1 for LL-RR, 9.5 for AA, and up to about 60 for LR, RL, VV, and LR+RL) is an improvement within a model that does not describe the data, not evidence for a nonzero ε. The most likely origin, that the statistical uncertainties alone do not represent the spread of the measurements, is the same systematic problem raised in the previous comment. The paper should report χ²_min/ndf and p-values for every model and address this issue directly.","section":"Appendix, Figures 1-9"},{"comment":"The manuscript never states per-model significances, and the abstract's 'over 3σ' claim is inconsistent with the paper's own numbers. The figure headers print the fit significance explicitly (labeled 'b' in units of σ, e.g., b = 3.07453 for AA in Figure 1, b = 7.73976 for VV in Figure 2, b = 7.70495 for LR in Figure 5, b = 0.32788 for LL-RR in Figure 9), and the ε/σ_ε ratios in Table 1 give the same values: six of the nine models (LL, RR, LR, RL, VV, LR+RL) deviate from zero at 6.2-7.7σ, while only AA is just above 3σ. The text reports none of these significances or any p-values, so the central claim cannot be checked from the table alone; conversely, if the fits were valid, the conclusion would be far stronger than the abstract's 'some derivation over 3σ'. In addition, no look-elsewhere penalty is applied for testing nine related models on the same 125 bins; because the models are strongly correlated the effective penalty is smaller than nine, but some penalty is required and none is given.","section":"Table 1, Figures 1-9, Abstract"}],"minor_comments":[{"comment":"'derivation' should be 'deviation', and the abstract as printed in the paper body omits the '(statistical uncertainty only)' qualification that appears in the arXiv metadata abstract and in Section 5, so the printed version states the conclusion more strongly than the analysis supports.","section":"Abstract"},{"comment":"The ε column contains apparent formatting artifacts (e.g., '0 .00446 ± 0.00071'), the column alignment is broken, and the relation between the model signs (η_LL, η_RR, η_LR, η_RL) and the sign of ε in Eq. (12) is never defined, so the reader cannot determine which sign choices were fitted.","section":"Table 1"},{"comment":"The symbol 'b' is used for two different quantities (the parabola center in the fit line and the significance in units of σ in the header line), the header text contains a doubled equals sign ('√1/a = = 0.00017'), and the text never defines a, b, and c, so the figures can only be read by reverse-engineering.","section":"Figures 1-9"},{"comment":"The data description is too coarse to reproduce the analysis: the paper does not list the OPAL energy points and cos θ binning that produce 105 bins, how the additional 20 DELPHI/L3 bins are defined, how the different acceptance cuts (acollinearity < 10° vs < 25°) are matched in the Monte Carlo, or how cross-experiment normalizations are handled.","section":"Sections 3 and 4"},{"comment":"The sentence 'the lowest order flavor-diagonal and helicity-conserving operators have dimension' is incomplete (the dimension, 6, is missing), and the assumed value of g²/4π in Eq. (12) is not stated in the main text, only in the unnumbered sentence preceding it.","section":"Section 2"},{"comment":"The paper does not quantitatively compare its AA-model result with the earlier hint of an axial-vector contact interaction reported by Bourilkov (Ref. [5]) from the same e+e- → e+e-(γ) data; such a comparison would be a natural consistency check, since Ref. [5] found a much weaker effect after a full treatment of systematics.","section":"Reference [5]"}],"recommendation":"reject","confidential_remarks":"The rejection rests on two self-contained facts visible in the manuscript itself: (1) the χ² used for the results contradicts the χ² defined in Eq. (13) by omitting σ_Sys, a point the authors disclose in Section 5; and (2) the best-fit models are rejected by their own χ² minima (χ²/ndf ≈ 4.2-4.7 in Figures 1-9), so the printed 3.1-7.7σ significances cannot be interpreted as evidence for ε ≠ 0. The manuscript is also extremely thin—roughly two pages of text, no data tables, no per-model p-values, no conclusion section—and does not meet the reporting standards of a journal article. Given the conflict with the published LEP results on the same data, I would not encourage a major revision of this draft; a resubmission would need the full published systematics with correlations, per-model p-values, a trials-factor treatment, and a resolution of the absolute χ² problem."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"What's new: this paper takes the standard LEP contact-interaction parameterization and fits it to a larger ensemble of published Bhabha data (OPAL, L3, DELPHI) across 130–207 GeV, covering nine chirality models. That is a legitimate extension of Bourilkov's earlier axial-vector hint [5] to a broader dataset and a fuller model scan. The use of Babayaga@NLO for the SM prediction is a reasonable choice, and the paper is transparent enough to show the fitted ε and Λ values with their parabolic uncertainties.\n\nThe core problem is the significance. Eq. (13) defines χ² with σ_Stat and σ_Sys, but Section 5 says the results only include statistical uncertainty. That is not a minor omission. LEP differential cross sections have correlated systematics from luminosity, acceptance, and radiative corrections that are at least comparable to the statistical errors in several bins. Ignoring them inflates any ε shift, and the abstract's own parenthetical \"(statistical uncertainty only)\" concedes the deviations are not robust. The paper also never reports χ²/ndf for the table's fits; the figures show χ²_min around 578 for 125 bins, i.e. χ²/ndf ≈ 4.7. A model that does not describe the data cannot give a reliable significance for its parameter. The quoted significances are further suspect because no look-elsewhere penalty is applied across the nine models, and the abstract says \"some derivation over 3σ\" while the table implies up to 8σ.\n\nWhat is good: the paper includes total cross sections where it finds no effect, which is an honest cross-check. The model scan is systematic, and the references to Bourilkov and earlier LEP contact-interaction analyses situate the work properly.\n\nNet: the central claim that Bhabha data prefer a non-zero contact interaction is unsupported. The analysis is a re-analysis of published data with a known method; the novelty is modest but present. The paper is not publishable as a research claim without a full treatment of systematics, bin correlations, fit quality, and trials factor. A serious editor could send it for review only if the authors were expected to fix these issues; as written, the statistical flaw is load-bearing. I would not cite it or bring it to a reading group, though it might be a useful negative example in a methods discussion.","headline":"A diligent but statistically flawed re-fit of LEP Bhabha data to contact-interaction models; the claimed deviations are not evidence once systematics and fit quality are taken into account.","tokens_in":5771,"tokens_out":5264,"would_cite":false,"duration_ms":42993,"reading_group":"no","serious_thinker":"no","would_accept_peer_review":false},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"deepseek-v4-flash","headline":"The paper claims that published LEP Bhabha-scattering data, fitted bin-by-bin in angle, prefer a nonzero four-fermion contact-interaction parameter with implied scale $\\Lambda\\approx13$–31 TeV, while total cross sections alone show no…","keywords":["contact interaction","Bhabha scattering","LEP","OPAL","four-fermion operators","compositeness","QED","beyond Standard Model"],"falsifier":"Recompute the same $\\chi^2$ for the same 125 bins and the same Babayaga@NLO prediction, now adding the published correlated systematic uncertainties of the OPAL, L3, and DELPHI measurements. If every fitted $\\varepsilon$ then lies within $2\\sigma$ of zero, the central claim is falsified; if at least one model still exceeds $3\\sigma$, the hint survives its weakest point.","tokens_in":4690,"feed_emoji":"⚛️","tokens_out":10688,"duration_ms":91795,"temperature":0.7,"pith_summary":"This paper tries to establish that published LEP Bhabha-scattering data, mainly from the OPAL detector with additional bins from L3 and DELPHI, contain a hint of a contact interaction beyond the Standard Model. The analysis writes the differential cross section as $\\mathrm{SM}(s,t)+\\varepsilon\\,C_{\\mathrm{Int}}(s,t)+\\varepsilon^2\\,C_{\\mathrm{CI}}(s,t)$, with $\\varepsilon=(g^2/4\\pi)\\,\\mathrm{sgn}(\\eta)/\\Lambda^2$, and fits $\\varepsilon$ to 125 angular bins. For total cross sections the fit finds no significant deviation from zero, but for differential cross sections several chirality models return $\\varepsilon$ values whose deviation from zero exceeds $3\\sigma$ when only statistical uncertainties are counted. The implied contact-interaction scale $\\Lambda$ is roughly 13–31 TeV depending on the model. If the hint is real, electrons are not point-like at very short distances and a new four-fermion interaction enters through interference with the Standard Model.","feed_headline":">3σ deviation in Bhabha data hints at contact interaction","feed_subtitle":"Fitting 125 angular bins at 130–207 GeV gives Λ≈13–31 TeV, but only statistical errors are counted.","key_machinery":"The load-bearing object is the signed parameter $\\varepsilon = (g^2/4\\pi)\\,\\mathrm{sgn}(\\eta)/\\Lambda^2$, which measures the inverse-square scale of a four-fermion contact interaction and fixes its sign. The machinery is a $\\chi^2$ fit over 125 differential bins in which the theoretical cross section is expanded to second order in $\\varepsilon$, separating the pure Standard Model term, the SM–contact-interference term, and the pure contact term. That separation is what allows a small $\\varepsilon$ to produce a visible deviation in the angular distribution while leaving the integrated rate close to the Standard Model prediction. The same machinery is applied to eight chirality models, differing only in which helicity couplings $\\eta_{LL}, \\eta_{RR}, \\eta_{LR}, \\eta_{RL}$ are switched on.","core_discovery":"The central claim is that the angular shape of $e^+e^-\\to e^+e^-(\\gamma)$ at LEP energies favours a nonzero contact-interaction parameter. Using the expansion $d\\sigma/d\\Omega = \\mathrm{SM}(s,t)+\\varepsilon\\,C_{\\mathrm{Int}}(s,t)+\\varepsilon^2\\,C_{\\mathrm{CI}}(s,t)$ and scanning $\\varepsilon$ against published bins, the authors obtain values such as $(0.00446\\pm0.00071)\\,\\mathrm{TeV}^{-2}$ for the LL model ($\\Lambda\\approx14.97$ TeV), $(0.00553\\pm0.00072)\\,\\mathrm{TeV}^{-2}$ for LR and RL ($\\Lambda\\approx13.44$ TeV), and $(0.00133\\pm0.00017)\\,\\mathrm{TeV}^{-2}$ for VV ($\\Lambda\\approx27.42$ TeV). The theoretical predictions come from the Babayaga@NLO generator in the same phase space as the measurements, and the paper states that the quoted $\\chi^2$ and significances include only statistical uncertainties. Because the total-cross-section fits show no significant effect, the claimed signal is entirely an angular-distribution effect.","pith_inferences":["The paper does not include systematic uncertainties; an editorially added test is to fold in the published correlated systematics of the OPAL, L3, and DELPHI bins, which could plausibly push every model below $3\\sigma$.","A quick diagnostic the paper does not report is which angular bins drive the $\\chi^2$; contact-interference terms in Bhabha scattering grow toward large $|\\cos\\theta|$, so a real signal should concentrate in the forward and backward bins.","Applying the same $\\varepsilon$-expansion to $e^+e^-\\to\\mu^+\\mu^-$ and $e^+e^-\\to\\tau^+\\tau^-$ LEP data would test whether the contact interaction is electron-specific or flavour-universal."],"forward_implications":["Future $e^+e^-$ analyses should fit angular bins rather than only total cross sections, because the claimed signal is invisible in the angle-integrated rate.","A real contact interaction at the fitted scales of 13–31 TeV would predict deviations that grow with centre-of-mass energy, making higher-energy $e^+e^-$ or muon colliders the natural place to confirm or exclude it.","The different fitted $\\Lambda$ values across the LL, RR, LR, RL, VV, AA, LL$\\pm$RR, and LR+RL models mean angular data can in principle distinguish which chiral operator structure is responsible.","If the hint is real, the earlier axial-vector contact-interaction hint is corroborated rather than appearing as a fluctuation in one data set."],"supporting_citations":[{"why":"Supplies the main OPAL differential cross-section data, 105 bins at 189–207 GeV with acollinearity below 10°, that drive the fit.","marker":"[1]"},{"why":"Adds DELPHI fermion-pair measurements in the angular range 44°–136° used as a second data set.","marker":"[2]"},{"why":"Adds L3 measurements of lepton-pair production at 130–189 GeV used as further input.","marker":"[3]"},{"why":"Provides the earlier OPAL four-fermion contact-interaction test at 130–140 GeV whose helicity-amplitude conventions the present analysis reuses.","marker":"[4]"},{"why":"Supplies the epsilon convention and the earlier axial-vector contact-interaction hint that this paper re-tests with more data.","marker":"[5]"},{"why":"Provides the Babayaga@NLO event generator used for the Standard Model predictions in the same phase space as the measured bins.","marker":"[6]"},{"why":"Supplies the effective four-fermion operator formalism and the contact-interaction amplitude structure used in the cross-section equations.","marker":"[8]"}],"fun_headline_variants":["Angular Bhabha data shows >3σ contact-interaction hint","LEP Bhabha angular bins suggest new contact interaction","Differential cross-section fit gives >3σ deviation at LEP","Bhabha angular shape hints at beyond-SM contact term (stat. only)","Contact interaction? >3σ in Bhabha differential data, stat. only"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The entire significance estimate assumes that the published experimental bins carry no systematic uncertainty, so that statistical errors alone are sufficient in the $\\chi^2$; if the real per-bin systematics are comparable to the statistical errors, the reported $>3\\sigma$ deviations could shrink and the contact-interaction claim would not follow.","fun_headline_variants_meta":{"raw":{"variants":["Angular Bhabha data shows >3σ contact-interaction hint","LEP Bhabha angular bins suggest new contact interaction","Differential cross-section fit gives >3σ deviation at LEP","Bhabha angular shape hints at beyond-SM contact term (stat. only)","Contact interaction? >3σ in Bhabha differential data, stat. only"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000197,"raw_usage":{"total_tokens":1329,"prompt_tokens":874,"completion_tokens":455,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":490,"completion_tokens_details":{"reasoning_tokens":359}},"tokens_in":490,"tokens_out":455,"duration_ms":4642,"temperature":1.0,"reasoning_tokens":359,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-12T14:00:57.893974+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Recompute the same $\\chi^2$ for the same 125 bins and the same Babayaga@NLO prediction, now adding the published correlated systematic uncertainties of the OPAL, L3, and DELPHI measurements. If every fitted $\\varepsilon$ then lies within $2\\sigma$ of zero, the central claim is falsified; if at least one model still exceeds $3\\sigma$, the hint survives its weakest point.","supporting_citations":[{"cited_title":"The European Physical Journal C, 33(2):173–212, March 2004","cited_arxiv_id":null,"evidence_quote":"Supplies the main OPAL differential cross-section data, 105 bins at 189–207 GeV with acollinearity below 10°, that drive the fit."},{"cited_title":"Measurement and interpretation of fermion- pair production at lep energies of 183 and 189 GeV","cited_arxiv_id":null,"evidence_quote":"Adds DELPHI fermion-pair measurements in the angular range 44°–136° used as a second data set."},{"cited_title":"Measurement of hadron and lepton-pair production at 130 < s <189GeV at lep","cited_arxiv_id":null,"evidence_quote":"Adds L3 measurements of lepton-pair production at 130–189 GeV used as further input."},{"cited_title":"Test of the four-fermion contact interaction in e+ e- collisions at 130–140 GeV","cited_arxiv_id":null,"evidence_quote":"Provides the earlier OPAL four-fermion contact-interaction test at 130–140 GeV whose helicity-amplitude conventions the present analysis reuses."},{"cited_title":"Bourilkov","cited_arxiv_id":null,"evidence_quote":"Supplies the epsilon convention and the earlier axial-vector contact-interaction hint that this paper re-tests with more data."},{"cited_title":"Status of the babayaga event generator","cited_arxiv_id":null,"evidence_quote":"Provides the Babayaga@NLO event generator used for the Standard Model predictions in the same phase space as the measured bins."},{"cited_title":"Eichten, Kenneth D","cited_arxiv_id":null,"evidence_quote":"Supplies the effective four-fermion operator formalism and the contact-interaction amplitude structure used in the cross-section equations."}],"review_version":1}