REVIEW 3 major objections 5 minor 23 references
New BaBar studies of high-order radiation and the new landscape of data-driven HVP predictions of the muon g-2
T0 review · 3 major / 5 minor · reviewed 2026-08-11 · deepseek-v4-flash
Pith's one-line read BaBar's first (N)NLO ISR radiation measurement shows that the Phokhara generator overproduces small-angle one-photon emission, implying hidden systematic errors in KLOE and BESIII that reshape the muon g-2 HVP landscape.
desk verdict A fair proceedings summary of already-published BaBar and HVP results; the generator claim is real and independently confirmed, but the KLOE/BESIII impact is not self-contained and rests entirely on Ref. [16]'s unseen fast simulations. 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 key object is the (N)NLO additional-radiation measurement: kinematic fits that count events with zero, one, or two extra photons above an energy threshold in e+e- -> mu+mu- gamma and e+e- -> pi+pi- gamma ISR data. This provides an in-situ test of two generators, Phokhara (full NLO matrix elements) and AfkQed (NLO plus NNLO in the collinear approximation), by comparing the rate and angular distribution of the extra photon or photons. The comparison exposes the generator discrepancy, and the fast simulations then translate that discrepancy into estimates of the systematic shift for the KLOE and BESIII event selections.
What would settle it
Compare Phokhara's predicted rate and angular distributions of small-angle one-photon ISR directly against KLOE or BESIII data events that pass their actual selection, before the Phokhara-based corrections are applied; if the data-to-MC ratio is consistent with unity there, the claimed systematic effect vanishes. Alternatively, reweight those data by AfkQed or an NNLO-capable generator and check whether the pi+pi- cross sections shift by the predicted amount.
Extended reading notes
Core claim
The central discovery, on the paper's own terms, is that the apparent muon g-2 HVP crisis is partly a Monte Carlo modeling effect. The BaBar measurement of additional radiation in initial-state-radiation events shows, for the first time at next-to- and next-to-next-to-leading order with one and two additional photons, that Phokhara's one-photon small-angle ISR rate is higher than in data and that NNLO contributions are missing from Phokhara, whereas AfkQed describes the data reasonably. The author infers, using fast simulations of the KLOE and BESIII analyses, that this generator bias propagates into their e+e- -> pi+pi-(gamma) cross sections as significant systematics; including KLOE in the dispersive average yields a muon g-2 prediction more than five sigma away from experiment, while excluding it leaves a 2.5-2.8 sigma difference and better agreement with lattice QCD.
Load-bearing premise
The estimate that KLOE and BESIII are significantly affected depends on fast simulations of their event selection and of how Phokhara is used in their corrections; if those simulations do not faithfully reproduce the real analyses, the claimed impact would not follow.
Editorial extensions
If this is right
- If the Phokhara bias is real, the KLOE and BESIII e+e- -> pi+pi- cross sections carry underestimated systematic uncertainties, which would explain their persistent pulls below the combined e+e- average.
- The dispersive HVP prediction for the muon g-2 depends strongly on which e+e- datasets are included: excluding KLOE leaves a 2.5-2.8 sigma difference with the direct measurement, while including it raises the tension beyond five sigma.
- The disagreement between data-driven and lattice QCD predictions of the muon g-2 is reduced when the apparently biased measurements are down-weighted, suggesting the gap is not primarily new physics but generator modeling.
- AfkQed's reasonable description of two-photon events suggests that ISR analyses should adopt generators with NNLO contributions or otherwise correct for missing higher-order radiation.
- Future e+e- measurements using loose selections, like the upcoming BaBar result without particle identification, can provide an independent check of the affected cross sections.
Reading between the lines
- Editorial inference: If the Phokhara bias is confirmed inside KLOE's own event selection, the KLOE cross sections are likely shifted in the rho region, and reweighting them upward would bring KLOE closer to BaBar and CMD-3 while moving the data-driven HVP value toward the lattice QCD result.
- Editorial inference: The same small-angle ISR mis-modeling could affect other ISR-based hadronic cross-section measurements beyond KLOE and BESIII, since many rely on Phokhara for corrections; the paper's call for generator tests naturally extends to any such analysis.
- Editorial inference: A direct test would be to re-analyze KLOE or BESIII data with AfkQed or an NNLO-capable generator instead of Phokhara; if the cross-section shift matches the fast-simulation estimate, the generator origin of the discrepancy would be established.
- Editorial inference: The paper's findings imply that the experimental uncertainty quoted for the muon g-2 prediction is not fully reliable until ISR generators are validated on measured multi-photon distributions, not just on total cross sections.
Editorial analysis
A structured set of objections, weighed in public.
Referee Report
Summary. This ICHEP2024 proceedings paper summarizes recent BaBar studies of additional (NLO and NNLO) photon radiation in e+e- -> mu+mu- gamma and e+e- -> pi+pi- gamma initial-state-radiation events, using the full BaBar data sample. The data are compared with the Phokhara and AfkQed Monte Carlo generators. The paper reports that Phokhara over-predicts the rate of small-angle one-photon ISR emission, while AfkQed describes the data reasonably well, and notes an independent Belle-II confirmation of the Phokhara discrepancy in the pi+pi-pi0 channel. It then argues that this generator discrepancy has negligible impact on BaBar's own pi+pi- cross-section measurement (which uses a loose inclusive selection) but could constitute a significant systematic effect for KLOE and BESIII, which select LO event topologies and rely on Phokhara for hard NLO corrections. On this basis, the paper revisits the data-driven dispersive prediction of the muon g-2 hadronic vacuum polarization, emphasizing tensions among KLOE, CMD-3, BaBar, tau data, lattice QCD, and the direct g-2 measurement.
Significance. If the central claim holds, the paper would identify a concrete, data-driven origin for part of the longstanding tension among e+e- -> pi+pi- cross-section measurements and would materially alter the interpretation of the data-driven HVP prediction of the muon g-2. The BaBar measurement of high-order radiation and the independent Belle-II confirmation of the Phokhara excess are important and credible contributions. However, the paper's impact statement depends on fast-simulation studies of KLOE and BESIII (Ref. [16]) that are not described in this manuscript, so the quantitative transfer from the BaBar generator-level observation to the HVP landscape is not auditable from the text. The proceedings format can cite prior work, but the abstract's claim that the impact was 'estimated and found to be indicative of significant systematic effects' is a load-bearing assertion that the body does not substantiate, making the significance conditional until that material is presented or the claim is softened.
major comments (3)
- [Abstract and Section 3] The abstract claims that the impact on the KLOE and BESIII measurements 'is estimated and found to be indicative of significant systematic effects', but Section 3 only states that these aspects 'have been further studied with fast simulations, questioning the KLOE systematic uncertainties [16]'. No quantitative result, simulation setup, or uncertainty treatment from Ref. [16] is presented in this paper. Because the title and Abstract frame the entire paper around this 'new landscape' of HVP predictions, the fast-simulation estimate is load-bearing; without at least a summary of its inputs, outputs, and assumptions, the reader cannot audit the claimed systematic effects or the significance statements in Section 4.
- [Abstract vs. body] The Abstract refers to 'the impact on the KLOE and BESIII measurements', but the body of the paper never mentions BESIII; Section 3 discusses only KLOE's 'LO' selection and Ref. [16]'s fast simulations, and Section 4 uses only KLOEwide and KLOEpeak variants. The Abstract therefore overstates the scope of the reported analysis. Either a BESIII-specific discussion should be added, or the Abstract should be narrowed to KLOE.
- [Section 4] The significance values quoted in Section 4 (e.g., '3.8σ above KLOEpeak', '2.9σ tension', '2.5σ (2.8σ)', 'larger than 5σ') are given without describing how the significances are computed, including how systematic correlations among the KLOE, BaBar, CMD-3, and tau inputs are propagated. These numbers are central to the paper's conclusion, so the procedure (or a precise pointer to equations in Ref. [16]) should be stated in the text, including the treatment of non-Gaussian systematic uncertainties mentioned in Section 5.
minor comments (5)
- [Section 3] The definition of 'NLO' and 'NNLO' depends on an unspecified additional-photon energy threshold ('having the energy above some given threshold'); the threshold value should be given or its definition in Ref. [15] reproduced.
- [Figure 1, bottom-right panel] The significance-of-difference calculation is said to account for correlations, but the method is not outlined; a reference or one-sentence description would aid the reader.
- [Section 4] The sentence 'Combining BaBar, CMD-3, tau (and BMW), a difference of 2.5σ (2.8σ) is found w.r.t. the experiment' is ambiguous about which combination yields which significance; please separate the statements.
- [References] Reference [20] is incomplete (no journal or publication status); please complete the citation.
- [Title and formatting] The title and text use non-standard glyphs ('𝒈 − 2', '𝜋+𝜋−') that may not render correctly in all formats; standard math fonts are recommended.
Circularity Check
No construction circularity: the BaBar/Phokhara discrepancy is a measured data-versus-MC comparison confirmed by Belle-II; only the KLOE/BESIII impact is imported from the author's own prior fast-simulation study, which is a self-reliance and auditability concern rather than a definitional or fitted-input circularity.
full rationale
The paper's central observational claim is that Phokhara over-predicts the rate of small-angle one-photon ISR radiation. This is established by comparing BaBar data with generator predictions, and it is independently confirmed by a Belle-II measurement of the pi+pi-pi0 channel. No parameter is fitted to the quantity being predicted, and the BaBar cross-section itself uses a loose selection that minimizes Monte Carlo dependence, so the generator discrepancy is not smuggled into the BaBar result. The HVP dispersive values are constructed from published external measurements (BaBar, CMD-3, KLOE, tau) using a combination tool described in the authors' prior work and validated by a closure test. The only notable self-reliance is that the quantitative estimate of the impact on KLOE and BESIII is delegated to Ref. [16], a fast-simulation study by the same group (including the author of these proceedings), rather than being re-derived or described here. That is a citation to a prior peer-reviewed analysis, not a reduction of the conclusion to its own input by construction; the underlying generator-level discrepancy is independently measurable and has been confirmed by another experiment. Therefore the derivation chain is not circular in the sense of fitted inputs being renamed as predictions or results being equivalent to their assumptions. Score 2 reflects the minor self-citation/self-reliance concerning the KLOE/BESIII impact, not an actual circularity.
Assumptions & free parameters
free parameters (1)
- Additional photon energy threshold for NLO/NNLO counting
assumptions (3)
- domain assumption The ISR method and the ratio technique (pion spectrum divided by muon spectrum) cancel most systematic uncertainties.
- domain assumption The Phokhara and AfkQed generators correctly implement the relevant QED matrix elements except for the identified NLO small-angle ISR rate issue.
- domain assumption Fast simulations of the KLOE and BESIII selections faithfully map the generator discrepancy into a cross-section bias.
Cite this review
Pith. "Pith review of New BaBar studies of high-order radiation and the new landscape of data-driven HVP predictions of the muon g-2." pith.science (2026). https://pith.science/paper/PEWO6W7Z
@misc{pith2026241211327,
author = {Pith},
title = {Pith review of: New BaBar studies of high-order radiation and the new landscape of data-driven HVP predictions of the muon g-2},
year = {2026},
howpublished = {\url{https://pith.science/paper/PEWO6W7Z}},
note = {Machine review of arXiv:2412.11327}
}
abstract
A measurement of additional radiation in $e^+e^- \to \mu^+\mu^- \gamma$ and $e^+e^- \to \pi^+\pi^- \gamma$ initial-state-radiation events is presented using the full $BaBar$ data sample. For the first time results are presented at next-to- and next-to-next-to-leading order, with one and two additional photons, respectively, for radiation from the initial and final states. The comparison with the predictions from Phokhara and AfkQed generators reveals discrepancies for the former in the one-photon rates and angular distributions. While this disagreement has a negligible effect on the $e^+e^- \to \pi^+\pi^- (\gamma)$ cross section measured by $BaBar$, the impact on the KLOE and BESIII measurements is estimated and found to be indicative of significant systematic effects. The findings shed a new light on the longstanding deviation among the muon $g-2$ measurement, the Standard Model prediction using the data-driven dispersive approach for calculation of the hadronic vacuum polarization (HVP), and the comparison with lattice QCD calculations.
Figures
Reference graph
Works this paper leans on
- [16]
-
[1]
1See also the contribution by Z
The CMD-2 Collaboration, JETP Lett.82(2005), 743-747 [arXiv:hep-ex/0603021]. 1See also the contribution by Z. Fodor at this conference, on a new result with lattice QCD (BMW) improvements and data-driven inputs (DMZ) at large-t [19]. 5 New BaBar studies of high-order radiation and the new landscape of data-driven HVP predictions of the muon𝑔− 2 Bogdan Malaescu
arXiv 2005
-
[2]
The CMD-2 Collaboration, JETP Lett.84 (2006), 413-417 [arXiv:hep-ex/0610016]
arXiv 2006
-
[3]
The CMD-2 Collaboration, Phys. Lett. B648 (2007), 28-38 [arXiv:hep-ex/0610021]
arXiv 2007
-
[4]
The SND Collaboration, J. Exp. Theor. Phys.103 (2006), 380-384 [arXiv:hep-ex/0605013]
arXiv 2006
-
[5]
The SND Collaboration, JHEP01(2021), 113 [arXiv:2004.00263]
arXiv 2021
-
[6]
The CMD-3 Collaboration, Phys. Rev. D109 (2024) no.11, 112002 [arXiv:2302.08834]
arXiv 2024
- [7]
Show all 23 references
-
[8]
The KLOE Collaboration, Phys. Lett. B670 (2009), 285-291 [arXiv:0809.3950]
2009 arXiv
-
[9]
The KLOE Collaboration, Phys. Lett. B700 (2011), 102-110 [arXiv:1006.5313]
2011 arXiv
-
[10]
The KLOE Collaboration, Phys. Lett. B720 (2013), 336-343 [arXiv:1212.4524]
2013 arXiv
-
[11]
The KLOE Collaboration, JHEP03(2018), 173, [arXiv:1711.03085]
2018 arXiv
-
[12]
The BaBar Collaboration, Phys. Rev. Lett.103 (2009), 231801 [arXiv:0908.3589]
2009 arXiv
-
[13]
The BaBar Collaboration, Phys. Rev. D86 (2012), 032013 [arXiv:1205.2228]
2012 arXiv
-
[14]
Davier, A
M. Davier, A. Hoecker, B. Malaescu and Z. Zhang, Eur. Phys. J. C80 (2020) no.3, 241 [arXiv:1908.00921]
2020 arXiv
-
[15]
The BaBar Collaboration, Phys. Rev. D108 (2023) no.11, L111103 [arXiv:2308.05233]
2023 arXiv
-
[17]
Borsanyi, Z
S. Borsanyi, Z. Fodor, J. N. Guenther, C. Hoelbling, S. D. Katz, L. Lellouch, T. Lip- pert, K. Miura, L. Parato and K. K. Szabo, et al. Nature 593 (2021) no.7857, 51-55 [arXiv:2002.12347]
2021 arXiv
-
[18]
Davier, Z
M. Davier, Z. Fodor, A. Gerardin, L. Lellouch, B. Malaescu, F. M. Stokes, K. K. Sz- abo, B. C. Toth, L. Varnhorst and Z. Zhang, Phys. Rev. D 109 (2024) no.7, 076019 [arXiv:2308.04221]
2024 arXiv
-
[19]
Lellouch and T
A.Boccaletti,S.Borsanyi,M.Davier,Z.Fodor,F.Frech,A.Gerardin,D.Giusti,A.Y.Kotov, L. Lellouch and T. Lippert,et al. [arXiv:2407.10913]
-
[20]
The Belle-II Collaboration, [arXiv:2404.04915]
-
[21]
M.Davier, A.Höcker, B.Malaescu, C.Z.YuanandZ.Zhang, Eur.Phys.J.C 74(2014)no.3, 2803 [arXiv:1312.1501]
2014 arXiv
-
[22]
131(2023)no.16,161802[arXiv:2308.06230]
TheMuong-2Collaboration,Phys.Rev.Lett. 131(2023)no.16,161802[arXiv:2308.06230]
2023
-
[23]
Muon 𝑔 − 2 Theory Initiative workshop Fermi- lab
Michel Davier, talk at the "Muon 𝑔 − 2 Theory Initiative workshop Fermi- lab", 2017, https://indico.fnal.gov/event/13795/session/10/contribution/ 47/material/slides/0.pdf. 6
2017
Reviewed August 11, 2026 · model on record in the stance chip above.
Discussion (0). Continue with ORCID to comment.