Pith. sign in

REVIEW 3 major objections 5 minor 20 references

Adding the ISR-shifted event energy to angular observables tightens the expected CEPC bound on the CP-odd Higgs coupling by about 20%.

Reviewed by Pith at T0; open to challenge. T0 means a machine referee read the full paper against a public rubric. the ladder, T0–T4 →

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗

T0 review · deepseek-v4-flash

2026-08-03 23:17 UTC pith:CPLXNVAO

load-bearing objection A genuinely new ISR-energy variable improves the CEPC Higgs CP limit by ~20% in a competent simulation, but the headline gain depends on an unreported κ_AZZ/Λ normalization, so treat the number as conditional. the 3 major comments →

arxiv 2511.06353 v2 pith:CPLXNVAO submitted 2025-11-09 hep-ph hep-ex

Testing Higgs CP properties at the CEPC with an additional ISR parameter

Alexey Drutskoy , Egor Vasenin This is my paper
classification hep-ph hep-ex PACS 12.60.-i13.66.-a14.60.St12.38.Qk
keywords Higgs CP propertiesCP violationinitial state radiatione+e− colliderCEPCHZZ couplingbinned likelihoodHiggs strahlung
verification ladder T0 review T1 audit T2 compute T3 formal T4 reserved

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

This paper proposes a new observable—the reconstructed total event energy E_RECO, which is lowered by initial-state radiation—as an additional handle on the CP properties of the Higgs boson in e+e−→HZ production at a 240 GeV lepton collider. Because ISR effectively scans the collision energy, and because the HZ cross section rises differently for CP-even (s-wave) and CP-odd (p-wave) amplitudes, the E_RECO distribution carries CP information beyond the three standard angles ϕ, θ1, θ2. Combining E_RECO with those angles in a binned likelihood at the CEPC with 5.6 ab−1 improves the expected 68% CL upper limit on the CP-odd coupling c~ZZ from 0.071 to 0.058, an improvement of roughly 20%. The analysis is carried out with a full Monte Carlo simulation including ISR and fast detector simulation, and the method is presented as a template for other Higgs CP measurements at future e+e− colliders.

Core claim

The central claim is that E_RECO is not merely a systematic nuisance but a discriminating variable: ISR reduces the effective center-of-mass energy, and the e+e−→HZ cross section σ(sqrt(s)) has a different shape for CP-even and CP-odd Higgs couplings (s-wave vs p-wave production). By binning the likelihood in E_RECO and re-optimizing selection cuts in each bin, the analysis extracts the CP information contained in the shape of the E_RECO spectrum. The result is an expected improvement of about 20% in the 1σ upper limit on the CP-odd coupling parameter c~ZZ, from 0.071 to 0.058, compared to using only the three angular observables.

What carries the argument

The key object is the E_RECO variable, defined as the sum of the reconstructed Higgs and Z energies, with the Higgs energy obtained from the recoil momentum under the assumption of a two-body final state. Its power comes from the relation between the cross section and the ISR-reduced effective energy: ISR smears the collision energy, and the CP-dependent s-wave (CP-even) vs p-wave (CP-odd) production amplitudes produce different E_RECO spectra. The analysis machinery is a four-bin binned likelihood over E_RECO, each bin with its own angular distributions and optimized selection requirements.

Load-bearing premise

The claimed improvement rests on the stated but unquantified choice that the CP-odd coupling strength κ_AZZ and scale Λ are tuned so that the total e+e−→HZ cross section at 240 GeV is independent of the mixing angle ψ_CP; if that normalization is changed, the E_RECO distributions for CP-even and CP-odd samples would differ by a different amount and the 20% gain could change or disappear.

What would settle it

Generate CP-even and CP-odd samples under a different physically motivated normalization of the CP-odd coupling (for instance, the one used in ECFA/FCC-ee studies) and repeat the multidimensional likelihood; if the E_RECO variable no longer improves the limit by ~20% (or worsens it), the claim is not robust. A direct analytic computation of the E_RECO spectrum difference under the paper's stated tuning would also settle the question.

Watch this falsifier — get emailed when new claim-graph text bears on it.

If this is right

  • If correct, the method shows that ISR-sensitive observables add complementary CP information, so future analyses at CEPC, FCC-ee, or ILC should include E_RECO-style variables.
  • The E_RECO distribution effectively performs an energy scan between the ZH threshold and √s; this could be exploited for other BSM studies that modify the energy dependence of the cross section.
  • The technique can be extended to other Higgs production channels (e.g., vector-boson fusion) and other decay modes.
  • The improvement implies that the total cross-section normalization independence of ψ_CP is a useful working point; the community may need to agree on a convention for the CP-odd coupling normalization to make such limits comparable.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • The magnitude of the improvement likely depends on the specific tuning of κ_AZZ and Λ that makes the total cross section at 240 GeV independent of ψ_CP; with a different normalization the E_RECO spectra could be more similar, and the 20% gain could shrink. This is testable by repeating the analysis with alternative coupling normalizations.
  • Because the cross-section difference between CP-even and CP-odd is largest near the ZH threshold, a collider operating closer to threshold (or a threshold scan) might see an even larger benefit from the E_RECO variable.
  • The method's reliance on MC templates for the expected distributions means systematic uncertainties in the ISR spectrum (e.g., the order of the ISR calculation) will directly feed into the CP limit; a data-driven calibration using e+e−→μ+μ−γ could stabilize the result.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit.

Referee Report

3 major / 5 minor

Summary. The paper proposes using the ISR-induced shift of the reconstructed total event energy E_RECO as an additional observable for measuring the CP properties of the HZZ coupling at the CEPC. The study uses WHIZARD with the Higgs Characterisation model, Pythia6, and Delphes with the CEPC detector card to simulate e+e- -> HZ, Z -> mu+mu- at sqrt(s)=240 GeV with 5.6 ab-1. Signal and the main backgrounds are generated; a binned likelihood in the three angular observables (phi, cos(theta1), cos(theta2)) is constructed separately in four E_RECO intervals. The authors claim an expected 68% CL upper limit on c~ZZ of 0.058, compared with 0.071 for an angular-only analysis, corresponding to about a 20% improvement (the abstract states 15%).

Significance. The idea that ISR-induced energy information can complement the standard angular observables is interesting and, if robust, would be a useful addition to the toolkit for CP measurements at future lepton colliders. The paper provides a concrete simulation chain including backgrounds, and the angular-only result is consistent with an earlier CEPC study. However, the quantitative central claim depends on an unspecified and unmotivated normalization of the HC couplings, and the exact statistical construction of the 'expected limit' is not described precisely enough to assess the quoted number. The improvement should therefore be treated as provisional pending clarification and robustness tests.

major comments (3)
  1. [Sec. II A and Eqs. (1), (2)] The central result is quoted in c~ZZ, but the paper never gives the relation between the HC parameters psi_CP, kappa_AZZ, Lambda and the conventional c~ZZ. The text says that kappa_AZZ and Lambda are 'chosen' so that the total cross section is independent of psi_CP at 240 GeV, but no numerical values or conversion formula are provided. This choice fixes the magnitude of the CP-odd amplitude and therefore controls the E_RECO shape difference that drives the claimed improvement. With a different normalization, the 20% improvement could change or disappear. Please specify the conversion between psi_CP and c~ZZ, state the chosen kappa_AZZ and Lambda values, and show how the limit depends on this choice.
  2. [Sec. III, Eq. (8)] The construction of the 'expected upper limit' is ambiguous. It is not stated whether N_i,j,k in the likelihood are taken from an Asimov dataset (the expected SM counts), from a single unweighted Monte Carlo realization, or from a pseudo-experiment ensemble. The sentence 'L1 are evaluated for simulated samples with c~ZZ values' is unclear about whether the data are fixed and only the expectation changes. If the quoted limit is from a single realization, it has a statistical spread that should be reported. Please specify the exact pseudo-experiment or Asimov procedure, and if appropriate report the median and an uncertainty band on the extracted limit.
  3. [Secs. II C and III] The claimed improvement is presented for a single analysis configuration, with no robustness checks. The result depends on the chosen E_RECO binning (four intervals with different p_mu+mu- cuts), the detector energy resolution, and the assumption of a particular ISR treatment, but none of these variations is explored. In addition, no systematic uncertainties are included or discussed. For a projection study this is acceptable as a first estimate, but the paper should at least state that the quoted limit is statistical only and provide some indication of how the 0.48 GeV energy resolution and the binning choices affect the 20% improvement.
minor comments (5)
  1. [Abstract vs. Sec. III] The abstract states the improvement is 15%, while Sec. III and the conclusions say about 20%. The quoted limits (0.071 to 0.058) imply (0.071-0.058)/0.071 ~ 18%. Please harmonize the numbers.
  2. [Sec. II C, Eq. (7)] The definition of E_RECO is called 'corrected for the ISR effects', but E_H is computed assuming zero total event momentum and using only the Z momentum. With a hard ISR photon this is not the true event energy. Please clarify the reconstruction procedure and the approximation being made.
  3. [Sec. III, Eq. (11)] The choice of the fit function ax^2+bx^4 is said to be motivated by |M|^2 being quadratic in c~ZZ. However, the test statistic is a full likelihood ratio with detector acceptance, backgrounds, and binning, not simply a matrix element squared. The polynomial should be treated as empirical or justified more carefully.
  4. [Table I] The formatting of Table I is corrupted in several places (e.g., '2114703.4' appears to be '21147 0.34' and '1992×10−4' is ambiguous). Please reformat the table so that columns and efficiencies are unambiguous.
  5. [General] The paper does not include a limitations paragraph. Since this is an MC projection, it would be helpful to state explicitly that no detector systematics, luminosity uncertainty, or ISR model uncertainty are included, and that the quoted limit is a statistical-only projection.

Circularity Check

0 steps flagged

No significant circularity: E_RECO sensitivity is a conditional Monte Carlo prediction under an explicit HC-model normalization, with no parameter fitted to the claimed limit and no load-bearing self-citation.

full rationale

Walk through the derivation chain. Samples are generated with WHIZARD/HC at fixed κ_HZZ=0 and with κ_AZZ/Λ chosen so the total e+e−→HZ cross section at √s=240 GeV is ψ_CP-independent (Sec. II A). This is an input modeling convention, not a fit to the target result. The E_RECO sensitivity follows from Eq. 5, σ(e+e−→HZ, √s, E_ISR)≈σ(e+e−→HZ, E_RECO,0), plus the computed energy dependence of σ for scalar, pseudoscalar, and mixed states (Fig. 3). The E_RECO distributions are calculated predictions of the model; they are not defined in terms of the claimed 0.071→0.058 improvement. The likelihood (Eq. 8), test statistic (Eq. 9), and polynomial fit are applied to simulated samples with and without E_RECO binning; the limit is extracted from the resulting curve, not imposed by an input. The angular-only result is cross-checked against the independent CEPC study [11], not a self-citation. The ad hoc κ_AZZ/Λ normalization is model-dependent — different choices would change the numerical improvement — but model dependence is not circularity, since the improvement is not forced by construction. The abstract/full-text discrepancy (15% vs 20%) is a reporting inconsistency, not a circular-derivation issue. Hence no circular step is identified.

Axiom & Free-Parameter Ledger

3 free parameters · 4 axioms · 0 invented entities

The central projection rests on several free choices: the CP-odd coupling normalization (κ_AZZ/Λ), the per-range p_μμ cuts, and the fitted polynomial coefficients. The E_RECO method also assumes a factorization of ISR from the CP-dependent cross section and that WHIZARD's ISR modeling is accurate. No invented entities are introduced.

free parameters (3)
  • κ_AZZ / Λ normalization = not quoted ('chosen such that total cross section independent of ψ_CP')
    Set in Sec. II A to equalize total cross sections at 240 GeV; this choice sets the size of the E_RECO spectral differences and therefore the claimed 20% improvement.
  • p_{μ+μ−} selection cuts per E_RECO range = [49–53] GeV (238–242), [48–51] (235–238), [39–49] (230–235), [26–40] (215–230)
    Hand-optimized separately in four E_RECO ranges (Sec. II D); the optimization procedure and validation are not described.
  • Polynomial coefficients a,b in S = a x^2 + b x^4 = not quoted
    Fitted to the simulated S values in Sec. III and used to extract the 68% CL limit; no uncertainties are reported.
axioms (4)
  • domain assumption The ISR spectrum factorizes from the CP-dependent cross section: σ(e+e−→HZ,√s,E_ISR) ≈ σ(e+e−→HZ,E_RECO,0) (Eq. 5).
    Load-bearing for treating E_RECO as an energy scan; assumes detector reconstruction cleanly maps E_RECO to effective √s.
  • domain assumption WHIZARD's all-order soft-photon ISR plus explicit hard-collinear ISR to O(α^3) accurately describes ISR at CEPC.
    The entire E_RECO distribution is generated by ISR; wrong ISR modeling would change the claimed improvement.
  • ad hoc to paper κ_AZZ and Λ can be tuned so that the total e+e−→HZ cross section at 240 GeV is independent of ψ_CP, with κ_HZZ=0.
    Invoked in Sec. II A to keep event counts equal; no numerical values or physical motivation are provided.
  • standard math S=−2ln(L1/L0) can be fitted by a x^2 + b x^4 and used as a χ²-like test statistic for 68% CL limits.
    Standard likelihood-ratio logic, but the paper does not verify the distributional assumption with toy experiments.

pith-pipeline@v1.3.0-alltime-deepseek · 7555 in / 16153 out tokens · 146001 ms · 2026-08-03T23:17:55.169990+00:00 · methodology

0 comments
read the original abstract

We evaluate the experimental sensitivity to the $CP$-odd admixture in the standard Higgs boson for the process $e^+e^-\to HZ$. The analysis is performed assuming the future lepton collider CEPC reference detector operating at $\sqrt{s}=240~\text{GeV}$ with statistics of $5.6~\text{ab}^{-1}$. Using the WHIZARD generator with the Higgs Characterisation model and the DELPHES detector simulation framework we obtain data samples for different $CP$-odd Higgs admixture parameters $\tilde{c}_{ZZ}$. The initial state radiation (ISR) effects are taken into account in WHIZARD. We develop a novel data-analysis method that exploits the ISR-induced shift in the reconstructed event energy $E_{\text{RECO}}$ as an additional source of information on the Higgs $CP$ structure. The method combines the ISR-sensitive event-energy information with three angular observables in a multidimensional likelihood analysis. The expected upper limit on the CP-odd Higgs admixture is improved by 15% compared with the standard analysis based on three angular variables.

Figures

Figures reproduced from arXiv: 2511.06353 by Alexey Drutskoy, Egor Vasenin.

Figure 1
Figure 1. Figure 1: FIG. 1: Kinematics for the [PITH_FULL_IMAGE:figures/full_fig_p002_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: FIG. 2: Distributions of the observables for the different Higgs boson [PITH_FULL_IMAGE:figures/full_fig_p003_2.png] view at source ↗
Figure 4
Figure 4. Figure 4: FIG. 4 [PITH_FULL_IMAGE:figures/full_fig_p004_4.png] view at source ↗
Figure 3
Figure 3. Figure 3: FIG. 3: Dependence of the cross section [PITH_FULL_IMAGE:figures/full_fig_p004_3.png] view at source ↗
Figure 5
Figure 5. Figure 5: FIG. 5 [PITH_FULL_IMAGE:figures/full_fig_p004_5.png] view at source ↗
Figure 6
Figure 6. Figure 6: FIG. 6: Distributions of the observables in the signal and background processes at [PITH_FULL_IMAGE:figures/full_fig_p005_6.png] view at source ↗
Figure 7
Figure 7. Figure 7: FIG. 7: Dependence of the test statistic [PITH_FULL_IMAGE:figures/full_fig_p007_7.png] view at source ↗

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

20 extracted references · 6 linked inside Pith

  1. [1]

    Chatrchyan et al

    S. Chatrchyan et al. (CMS Collaboration), Physics Let- ters B716, 1 (2012)

    2012

  2. [2]

    Aad et al

    G. Aad et al. (ATLAS Collaboration), Eur. Phys. J. C 83(2023) 563

    2023

  3. [3]

    The CEPC Study Group, arXiv:2312.14363 [physics.acc- ph]

    arXiv

  4. [4]

    Abada et al

    A. Abada et al. (FCC-ee Collaboration), Eur. Phys. J. Spec. Top.228, 261–623 (2019)

    2019

  5. [5]

    C.Adolphsenetal.(ILCCollaboration), arXiv:1306.6328 [physics.acc-ph]

    Pith/arXiv arXiv

  6. [6]

    Barklow, J

    T. Barklow, J. Brau, K. Fujii, J. Gao, J. List, N. Walker, K. Yokoya, arXiv:1506.07830 [hep-ex]

    Pith/arXiv arXiv

  7. [7]

    Aryshev et al

    A. Aryshev et al. (ILC Collaboration), arXiv:2203.07622 [physics.acc-ph]

    arXiv

  8. [8]

    Aicheler, P.N

    M. Aicheler, P.N. Burrows, N. Catalan, R. Corsini, M. Draper, J. Osborne, D. Schulte, S. Stapnes, M.J. Stu- art, arXiv:1903.08655 [physics.acc-ph]

    Pith/arXiv arXiv 1903

  9. [9]

    D. M. Asner et al. (ILC Collaboration), arXiv:1310.0763 [hep-ph]

    Pith/arXiv arXiv

  10. [10]

    Jeans and G

    D. Jeans and G. W. Wilson, Phys. Rev. D98, 013007 (2018)

    2018

  11. [11]

    Q. Sha, A. Fadol, F. Guo, G. Li, Y. Fang, J. Gu and X. Lou, Eur. Phys. J. C82, 981 (2022)

    2022

  12. [12]

    Cheung, Y.-n

    K. Cheung, Y.-n. Mao, S. Moretti, R. Zhang, Eur. Phys. J. C85(2025) 6, 700

    2025

  13. [13]

    ATLAS Collaboration, ATL-PHYS-PUB-2013-013

    2013

  14. [14]

    ECFA Collaboration, CERN-2025-005

    2025

  15. [15]

    Blondel, C

    A. Blondel, C. Grojean, P. Janot and G. Wilkinson, arXiv:2412.13130 [hep-ph]

    Pith/arXiv arXiv

  16. [16]

    Artoisenet, P

    P. Artoisenet, P. de Aquino, F. Demartin, R. Frederix, S. Frixione, F. Maltoni, M. K. Mandal, P. Mathews, K. Mawatari, V. Ravindran, et al., J. High Energ. Phys. 2013, 43 (2013)

    2013

  17. [17]

    Kilian, T

    W. Kilian, T. Ohl, J. Reuter, Eur.Phys.J.C71(2011) 1742

    2011

  18. [18]

    Moretti, T

    M. Moretti, T. Ohl, J. Reuter, arXiv: hep-ph/0102195

    Pith/arXiv arXiv

  19. [19]

    Sjöstrand, L

    T. Sjöstrand, L. Lönnblad, and S. Mrenna, arXiv:hep- ph/0108264

    arXiv

  20. [20]

    de Favereau, C

    J. de Favereau, C. Delaereet al.(The DELPHES 3 col- laboration), J. High Energy Phys. 02 (2014) 057

    2014