REVIEW 3 major objections 6 minor 8 references
New test on contact interactions in the data at $\sqrt{s}$ 130-207GeV by Bhabha scattering process $e^+e^-\to e^+e^-$
T0 review · 3 major / 6 minor · reviewed 2026-08-12 · deepseek-v4-flash
Pith's one-line read 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…
desk verdict 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. read the letter →
The pith
A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.
The reading
What carries the argument
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.
What would settle it
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.
Extended reading notes
Core claim
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.
Load-bearing premise
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.
Editorial extensions
If this is right
- 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.
Reading between the lines
- 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.
Editorial analysis
A structured set of objections, weighed in public.
Referee Report
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.
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 (3)
- [Section 5 vs Eq. (13)] 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.
- [Appendix, Figures 1-9] 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.
- [Table 1, Figures 1-9, Abstract] 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.
minor comments (6)
- [Abstract] '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.
- [Table 1] 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.
- [Figures 1-9] 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.
- [Sections 3 and 4] 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 2] 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.
- [Reference [5]] 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.
Circularity Check
No internal circularity: the epsilon fit is a standard parameter estimation against external published data; the statistical-only caveat is a robustness weakness, not a definitional circularity.
full rationale
The derivation chain is: contact-interaction Lagrangian -> differential cross section Eq. (2)-(5) with model coefficients -> parametrization Eq. (11) dsigma/dOmega = SM + epsilon*CInt + epsilon^2*CCI -> chi^2 in Eq. (13) against published OPAL/DELPHI/L3 differential cross sections -> best-fit epsilon and Lambda in Table 1. The parameter epsilon is a free theoretical parameter defined in Eq. (12) from the model scale and coupling, not constructed from the measured deviations. The best-fit epsilon values and their statistical errors are obtained by minimizing Eq. (13) over the same external data; this is standard fitting, and the reported '3-sigma deviation' is the statistical significance of the fitted epsilon relative to zero, not an out-of-sample prediction. No load-bearing step is equivalent to its input by construction. The paper contains no self-citations that carry the argument. The omission of systematics in Section 5 ('For now the results only includes statistical uncertainty') weakens the quoted significance, but that is a statistical robustness issue, not circularity. Likewise, converting fitted epsilon to Lambda uses the definitional relation epsilon = (g^2/4*pi) * sgn(eta)/Lambda^2, which is not a circular derivation.
Assumptions & free parameters
free parameters (9)
- ε (LL model) =
0.00446 ± 0.00071 TeV^-2
- ε (RR model) =
0.00456 ± 0.00073 TeV^-2
- ε (LR model) =
0.00553 ± 0.00072 TeV^-2
- ε (RL model) =
0.00553 ± 0.00072 TeV^-2
- ε (VV model) =
0.00133 ± 0.00017 TeV^-2
- ε (AA model) =
0.00139 ± 0.00045 TeV^-2
- ε (LL-RR model) =
0.00034 ± 0.00103 TeV^-2
- ε (LL+RR model) =
0.00225 ± 0.00103 TeV^-2
- ε (LR+RL model) =
0.00309 ± 0.00040 TeV^-2
assumptions (5)
- domain assumption Contact-interaction parameterization (eq. 2-12) is the correct effective field theory for new physics.
- domain assumption Babayaga@NLO provides an accurate Standard Model prediction for Bhabha scattering at LEP energies.
- domain assumption Individual bins are statistically independent and uncertainties are Gaussian.
- ad hoc to paper Systematic uncertainties can be neglected.
- ad hoc to paper No trials factor is needed when testing nine models and many bins.
Cite this review
Pith. "Pith review of New test on contact interactions in the data at $\sqrt{s}$ 130-207GeV by Bhabha scattering process $e^+e^-\to e^+e^-$." pith.science (2026). https://pith.science/paper/QXAMSDH2
@misc{pith2026241115686,
author = {Pith},
title = {Pith review of: New test on contact interactions in the data at $\sqrts$ 130-207GeV by Bhabha scattering process $e^+e^-\to e^+e^-$},
year = {2026},
howpublished = {\url{https://pith.science/paper/QXAMSDH2}},
note = {Machine review of arXiv:2411.15686}
}
abstract
We used data mainly collected by OPAL experiment to test the signal significance of a parameter $\varepsilon$, which is equal to zero in SM. For total cross sections, we obtained no derivation of enough significance. But for differential cross sections, some derivation over 3$\sigma$ are observed(statistical uncertainty only), indicating that this contact interaction might exist.
Figures
Figures from the paper (6 more)
Reference graph
Works this paper leans on
- [5]
-
[1]
The European Physical Journal C, 33(2):173–212, March 2004
Tests of the standard model and constraints on new physics from measurements of fermion- pair production at 189-209 GeV at lep. The European Physical Journal C, 33(2):173–212, March 2004
work page 2004
-
[2]
Measurement and interpretation of fermion- pair production at lep energies of 183 and 189 GeV
Paulo Abreu, W Adam, T Adye, P Adzic, Z Albrecht, T Alderweireld, GD Alekseev, R Alemany, T Allmendinger, PP Allport, et al. Measurement and interpretation of fermion- pair production at lep energies of 183 and 189 GeV. Physics Letters B, 485(1-3):45–61, 2000
work page 2000
-
[3]
Measurement of hadron and lepton-pair production at 130 < s <189GeV at lep
M Acciarri, P Achard, Oscar Adriani, M Aguilar-Benitez, J Alcaraz, G Alemanni, J Allaby, Alberto Aloisio, MG Alviggi, G Ambrosi, et al. Measurement of hadron and lepton-pair production at 130 < s <189GeV at lep. Physics Letters B, 479(1-3):101–117, 2000
work page 2000
-
[4]
Test of the four-fermion contact interaction in e+ e- collisions at 130–140 GeV
Gideon Alexander, J Allison, N Altekamp, K Ametewee, KJ Anderson, S Anderson, S Ar- celli, S Asai, D Axen, G Azuelos, et al. Test of the four-fermion contact interaction in e+ e- collisions at 130–140 GeV. Physics Letters B, 387(2):432–442, 1996
work page 1996
-
[6]
Status of the babayaga event generator
Carlo M Carloni Calame, Guido Montagna, Oreste Nicrosini, and Fulvio Piccinini. Status of the babayaga event generator. In EPJ Web of Conferences, volume 218, page 07004. EDP Sciences, 2019
work page 2019
-
[7]
Opal Collaboration and K Ackerstaff. Tests of the standard model and constraints on new physics from measurements of fermion-pair production at 130–172 GeV at lep. The European Physical Journal C-Particles and Fields, 2
-
[8]
Estia J. Eichten, Kenneth D. Lane, and Michael E. Peskin. New tests for quark and lepton substructure. Phys. Rev. Lett., 50:811–814, Mar 1983. 9
work page 1983
Reviewed August 12, 2026 · model on record in the stance chip above.
Discussion (0). Continue with ORCID to comment.