Pith. sign in

REVIEW 2 major objections 4 minor 45 references

A reanalysis shows the published four-lepton signal efficiency exceeds a mass-independent upper bound, weakening doubly charged Higgs mass limits by about 100 GeV.

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 →

T0 review · deepseek-v4-flash

2026-08-05 04:30 UTC pith:WCWKJO44

load-bearing objection A clean analytic upper bound on the four-lepton efficiency, and a plausible charge that ATLAS overestimates it—but the entire case rests on an unproven reading of the auxiliary cutflow's 'Yield' row. the 2 major comments →

arxiv 2608.03988 v1 pith:WCWKJO44 submitted 2026-08-04 hep-ph hep-ex

Revised exclusion limits on doubly charged Higgs bosons from a reanalysis of the ATLAS multi-lepton search at sqrt{s} = 13,TeV

classification hep-ph hep-ex
keywords doubly charged Higgs bosonfour-lepton signal efficiencyequal-branching-ratio benchmarktype-II seesawZee-Babu modelsignal cutflow consistencyCLs exclusion limittau leptonic decays
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 reexamines the four-lepton channel of the published search for pair-produced doubly charged Higgs bosons—the channel that drives the strongest mass limits on these hypothetical states. It argues that the signal efficiency used for that channel is too high: under the search's own equal-branching-ratio assumption, at most 29% of produced pairs can yield four reconstructed electrons or muons, yet the auxiliary cutflow implies retained fractions of 43–46%. The authors regenerate the signal independently, obtain yields 36–40% of the published ones across all tested masses, and recompute the 95% CL exclusion with the same statistical procedure, finding expected mass bounds about 100 GeV lower: 1065 to ~950 GeV in one benchmark and 880 to ~770 GeV in another. The result matters because these published bounds are widely used to constrain extensions of the Standard Model that contain doubly charged scalars.

Core claim

The central claim is that the four-lepton signal efficiency underlying the published search exceeds a strict, mass-independent ceiling. One doubly charged scalar arm yields two light leptons with probability 0.64, so the truth-level four-light-lepton fraction is at most 0.41; after tight-lepton efficiencies 0.88 (electron) and 0.95 (muon) it falls to 0.29. The search's auxiliary cutflow implies retained fractions of 0.43–0.46, above even the truth-level bound, and the paper shows hadronic tau or jet misidentification cannot close the gap without fake rates near 55%. Its independent simulation gives retained fractions consistent with 0.29 and an Ours/ATLAS yield ratio of 0.36–0.40. Recomputin

What carries the argument

The load-bearing object is the per-arm probability Σf2 = 0.64 that one doubly charged scalar decays into two same-sign light leptons, assembled from the six equal branching ratios (1/6 each) and the leptonic tau branching fractions (0.35 combined). Its square, P4^truth = 0.41, and the efficiency-adjusted value P4^reco = 0.29, form a mass-independent ceiling on any four-lepton signal selection, because both arms of the pair must each supply two light leptons. The ceiling is the instrument that exposes the inconsistency: dividing the auxiliary cutflow entries by the total yield gives fractions above 0.41, and the near-constant Ours/ATLAS ratio of 0.36–0.40 then turns that excess into evidence

Load-bearing premise

The comparison rests on taking the 'Yield' row of the published cutflow table as the total number of generated pp to H++ H-- events; if that row is already a prefiltered subset of events, the claimed inconsistency in the four-lepton efficiencies disappears.

What would settle it

Look at the generated-event count behind the published cutflow. If the 'Yield' row of the auxiliary table is the full number of pp to H++ H-- events generated for each mass point, then the paper's efficiency comparison is airtight; if that row already reflects a prefilter such as a generator-level or two-lepton requirement, the retained fractions are not global efficiencies and the claimed contradiction disappears. The experiment could settle this by releasing per-stage generator-level event counts and the filter definitions.

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

Share X Bluesky LinkedIn Reddit HN

If this is right

  • If the bound is correct, the published four-tight-lepton fractions of 0.43–0.46 cannot represent genuine signal from pair production under the equal-branching-ratio benchmark; the physical maximum before kinematic cuts is 0.29.
  • Corrected signal yields place the expected 95% CL cross-section limit roughly a factor of two above the published expected limit over the whole 400–1300 GeV range.
  • The expected lower mass bound for the left-right symmetric type-II seesaw doubly charged scalar falls from 1065 GeV to about 950 GeV.
  • The expected lower mass bound for the Zee–Babu doubly charged scalar falls from 880 GeV to about 770 GeV.
  • The flat 0.36–0.40 Ours/ATLAS yield ratio across mass points indicates a normalisation offset in the published four-lepton sample rather than a mass-dependent physics effect.

Where Pith is reading between the lines

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

  • The inconsistency could be settled directly if the experimental collaboration released the generated-event counts behind its cutflow table: if the 'Yield' row is already a filtered subset, the analytic bound does not apply, while if it is the full generated sample, the published efficiency stands contradicted.
  • The same per-arm bound can serve as a cheap sanity check for any future search or recast: under the equal-branching-ratio leptonic benchmark, no four-lepton efficiency above 29% is possible unless fake leptons or hadronic tau decays explicitly enter the selection.
  • If the normalisation offset is common to the signal samples, the two- and three-lepton signal regions of the same search should also show weaker limits once absolute yields are corrected; checking that would localise the discrepancy or confirm it is global.
  • Model projections for future colliders that reuse the published signal efficiencies would inherit the same overestimate; applying the bound as a pre-filter would make such projections more reliable.

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

2 major / 4 minor

Summary. The paper reexamines the ATLAS search for pair-produced doubly charged Higgs bosons in multi-lepton final states (ATLAS EXOT-2018-34, Eur. Phys. J. C 83 (2023) 605). Under the equal-branching-ratio benchmark used by ATLAS, the authors derive an analytic upper bound on the fraction of pp→H++H-- events that can yield four reconstructed light leptons: 0.41 at truth level and 0.29 after applying quoted electron and muon reconstruction efficiencies (Eqs. 9 and 10). They compare these bounds with the four-lepton cutflow entries of ATLAS auxiliary Table 5, obtaining retained fractions of 0.43–0.46, and argue that this inconsistency cannot be explained by hadronic-tau or jet fakes. They then generate the signal independently with MadGraph5+Delphes, obtain four-lepton yields about a factor 0.36–0.40 of the ATLAS yields, and recompute 95% CL expected limits using pyhf with ATLAS background predictions. The revised limits are claimed to be weaker by roughly a factor of two in cross section, shifting the expected mass bounds from 1065 GeV to ~950 GeV (left-right symmetric type-II seesaw) and from 880 GeV to ~770 GeV (Zee–Babu).

Significance. If the central comparison is valid, this is an important correction to a headline BSM exclusion limit: it would imply that the ATLAS four-lepton signal efficiency is overestimated and that the published mass limits are too aggressive by about 100 GeV. The analytic bound in Sec. II is a clean, parameter-free derivation that does not depend on detector simulation, and the paper appropriately adopts ATLAS background predictions rather than re-deriving them. The use of pyhf and the transparent CL_s procedure are also strengths. However, the significance is conditional on one essential premise: that the 'Yield' row in ATLAS auxiliary Table 5 represents the total number of produced pp→H++H-- events. The manuscript does not provide evidence for this identification, and if 'Yield' is a prefiltered sample the claimed inconsistency disappears. The revised limits additionally depend on an unprovided Delphes tuning. The paper is therefore of high potential interest, but its central claim is currently not fully established.

major comments (2)
  1. [Sec. II.D, Eq. (12), Table II] The central comparison assumes that the 'Yield' row of ATLAS auxiliary Table 5 is the total number of generated pp→H++H-- events (i.e., σ×L before any selection). The manuscript does not establish this. Auxiliary Table 5 is described only as 'normalised to 139 fb^-1', which is equally compatible with a prefiltered sample (for example, events passing a generator-level or two-lepton requirement). If 'Yield' is prefiltered, the retained fractions 0.43–0.46 are not bounded by P_truth4=0.41 or P_reco4=0.29, and the claimed inconsistency vanishes. Table II itself normalises the authors' simulation to the ATLAS Yield (Ratio=1.00 by construction), so the Ours/ATLAS ratios do not independently test the denominator. The authors must either demonstrate that the ATLAS Yield equals their σ×L, for instance by comparing it with the production cross-section times 139 fb^-1, or provide direct evidence fr
  2. [Sec. III and Sec. IV] The revised exclusion limits in Sec. IV rest on a Delphes 3 simulation with 'a card tuned to reproduce the ATLAS lepton reconstruction and identification efficiencies'. The card is not provided, the tuning is not quantified, and no closure test against the ATLAS cutflow is shown. Since the expected limits of ~950 GeV and ~770 GeV are among the paper's primary results, this is a reproducibility gap. Please provide the Delphes card and a validation table comparing per-lepton efficiencies, pT thresholds, and isolation/identification modelling with the cited ATLAS performance references. This is needed for the quantitative limit revision, although the analytic bound in Sec. II is independent of the simulation.
minor comments (4)
  1. [Sec. II.D, Table II] The sentence 'This ratio is numerically close to, but systematically below, P_truth4 = 0.41' is misleading. The Ours/ATLAS ratio is a ratio of selection efficiencies relative to the same (unverified) Yield denominator, not an absolute fraction of produced events. It does not by itself indicate a normalisation offset.
  2. [Sec. II.A, Eqs. (5)–(7)] The notation f(ee), f(eμ), f(μμ) does not explicitly state that the lepton pair is same-sign within each decay arm (both e+ in H++ decay and both e− in H−− decay). Clarify the sign convention to avoid confusion.
  3. [Sec. IV] The abstract says the expected limit lies 'roughly a factor of two or more' above ATLAS. In the text this factor varies with mass. Please state the mass range over which this factor holds and confirm that it refers to the cross-section upper limit, not to the mass exclusion.
  4. [Sec. II.D and Table II] The authors do not reproduce the relevant rows of ATLAS auxiliary Table 5 in the manuscript. A small table quoting the ATLAS 'Yield', 'four loose leptons', and 'four tight leptons' entries for the four benchmark masses would help readers verify the 0.50–0.51 and 0.43–0.46 fractions without consulting the auxiliary material.

Circularity Check

0 steps flagged

No significant circularity: the analytic bound and revised limits are derived from external inputs; self-citations are non-central.

full rationale

The paper's central chain is not circular. The four-lepton bound P_reco4 = 0.29 (Eq. 10) is computed from the equal-branching-ratio assumption (Eq. 1), PDG tau branching fractions (Eqs. 2–4), and ATLAS-quoted lepton efficiencies; none of these inputs is the ATLAS four-lepton signal efficiency or the ATLAS mass limit. The comparison with ATLAS auxiliary Table 5 is an external-data test, not a tautology. The signal regeneration in Sec. III uses MadGraph/Pythia/Delphes with ATLAS performance maps and is benchmarked against the cutflow, but the 'Ours/ATLAS' ratio is normalized to the same ATLAS Yield, making it an efficiency comparison rather than a fit to the four-lepton yield. The limit recomputation in Sec. IV adopts ATLAS background predictions and the same pyhf CL_s procedure; this is a conditional recomputation (if signal is reduced, limits weaken) and does not presuppose the conclusion. Self-citations [22,25,26] are phenomenological context, not load-bearing. The main substantive caveat — whether the ATLAS auxiliary 'Yield' row represents total production or a prefiltered sample — is a question about the external data's meaning, not a circularity in the authors' derivation; if 'Yield' is prefiltered, the claimed inconsistency may vanish, but that is a correctness risk, not a self-referential reduction. Therefore the circularity score is low.

Axiom & Free-Parameter Ledger

1 free parameters · 6 axioms · 0 invented entities

The central claim rests on external branching fractions, quoted ATLAS efficiencies, and the interpretation of the auxiliary cutflow. No new particles or mediators are introduced. The Delphes tuning is the main set of unprovided parameters affecting the revised limits.

free parameters (1)
  • Delphes detector-card tuning constants = not provided
    In Sec. III the Delphes card is 'tuned to reproduce' ATLAS lepton efficiencies and isolation; the tuning values are not reported and directly control the corrected signal yields used for the revised limits.
axioms (6)
  • domain assumption Equal branching ratio: B(H++ -> e+e+) = ... = 1/6 (Eq. 1).
    Adopted from the ATLAS benchmark; the derived bound uses it as input. If the true branching ratios are not equal, the ceiling changes.
  • domain assumption Tau leptonic branching fractions b_e = 0.18, b_mu = 0.17, b_l = 0.35 (Eqs. 2-4).
    External PDG values used in the calculation of the four-lepton bound.
  • domain assumption Tight-lepton efficiencies eps_e = 0.88 and eps_mu = 0.95 quoted from Ref. [27] are appropriate before kinematic acceptance.
    Used to reduce P_truth to P_reco; if these efficiencies are not per-lepton in the relevant phase space, the 0.29 bound changes, though it cannot exceed 0.41.
  • domain assumption The 'Yield' row of ATLAS auxiliary Table 5 counts all generated pp -> H++ H-- events.
    Needed for the central comparison in Sec. II D; not stated explicitly in the paper.
  • ad hoc to paper Delphes fast simulation with a tuned card reproduces ATLAS full-simulation signal efficiencies.
    Used for corrected yields in Secs. III-IV; no quantitative closure test is provided.
  • domain assumption ATLAS post-fit background predictions and uncertainties are taken as correct inputs to the limit.
    Adopted in Sec. IV; the paper only changes the signal side.

pith-pipeline@v1.3.0-daily-deepseek · 11464 in / 25386 out tokens · 299174 ms · 2026-08-05T04:30:16.151997+00:00 · methodology

0 comments
Cite this review

Pith. "Pith review of Revised exclusion limits on doubly charged Higgs bosons from a reanalysis of the ATLAS multi-lepton search at $\sqrt{s} = 13$,TeV." pith.science (2026). https://pith.science/paper/WCWKJO44

@misc{pith2026260803988,
  author       = {Pith},
  title        = {Pith review of: Revised exclusion limits on doubly charged Higgs bosons from a reanalysis of the ATLAS multi-lepton search at $\sqrts = 13$,TeV},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/WCWKJO44}},
  note         = {Machine review of arXiv:2608.03988}
}
Share X Bluesky LinkedIn Reddit HN
read the original abstract

The ATLAS search for pair-produced doubly charged Higgs bosons in multi-lepton final states using the full Run 2 dataset [Eur. Phys. J. C 83 (2023) 605] reports the strongest limits to date on the mass of doubly charged scalars, driven by an essentially background-free four-lepton channel. We show that the four-lepton signal efficiency implied by the auxiliary cutflow of that analysis exceeds a strict, mass-independent upper bound derived from the equal-branching-ratio assumption of the search, the leptonic $\tau$ branching fractions, and the ATLAS lepton reconstruction efficiencies. We show that this excess cannot be explained by hadronic $\tau$ or jet misidentification without invoking fake rates far above realistic values. We regenerate the signal independently and recompute the exclusion limit, using the corrected signal yields, the ATLAS background predictions and uncertainties, and the same $CL_s$ procedure implemented in pyhf. The resulting expected limit lies systematically above the ATLAS expected limit, by roughly a factor of two or more. This shifts the expected lower mass bound from $1065$ GeV to $\sim950$ GeV in the left-right symmetric type-II seesaw model, and from $880$ GeV to $\sim770$ GeV in the Zee--Babu model.

Figures

Figures reproduced from arXiv: 2608.03988 by Arun Kumar Nayak, Debabrata Sahoo, Kirtiman Ghosh.

Figure 1
Figure 1. Figure 1: FIG. 1. Distributions of the leading same-sign dilepton in [PITH_FULL_IMAGE:figures/full_fig_p005_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: FIG. 2. Distribution of the leading same-sign dilepton in [PITH_FULL_IMAGE:figures/full_fig_p006_2.png] view at source ↗
Figure 3
Figure 3. Figure 3: FIG. 3. Event yield in the four-lepton signal region (SR4L). [PITH_FULL_IMAGE:figures/full_fig_p006_3.png] view at source ↗
Figure 4
Figure 4. Figure 4: FIG. 4. Expected 95% CL upper limit on the [PITH_FULL_IMAGE:figures/full_fig_p007_4.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

45 extracted references · 19 canonical work pages · 9 internal anchors

  1. [1]

    Magg and C

    M. Magg and C. Wetterich, Phys. Lett. B94, 61 (1980)

  2. [2]

    Schechter and J

    J. Schechter and J. W. F. Valle, Phys. Rev. D22, 2227 (1980)

  3. [3]

    T. P. Cheng and L.-F. Li, Phys. Rev. D22, 2860 (1980)

  4. [4]

    Lazarides, Q

    G. Lazarides, Q. Shafi, and C. Wetterich, Nucl. Phys. B 181, 287 (1981)

  5. [5]

    R. N. Mohapatra and G. Senjanovic, Phys. Rev. D23, 165 (1981)

  6. [6]

    J. C. Pati and A. Salam, Phys. Rev. D10, 275 (1974), [Erratum: Phys.Rev.D 11, 703–703 (1975)]

  7. [7]

    R. N. Mohapatra and J. C. Pati, Phys. Rev. D11, 566 (1975)

  8. [8]

    Senjanovic and R

    G. Senjanovic and R. N. Mohapatra, Phys. Rev. D12, 1502 (1975)

  9. [9]

    Zee, Nucl

    A. Zee, Nucl. Phys. B264, 99 (1986)

  10. [10]

    K. S. Babu, Phys. Lett. B203, 132 (1988)

  11. [11]

    Georgi and M

    H. Georgi and M. Machacek, Nucl. Phys. B262, 463 (1985)

  12. [12]

    M. S. Chanowitz and M. Golden, Phys. Lett. B165, 105 (1985)

  13. [13]

    Singer, J

    M. Singer, J. W. F. Valle, and J. Schechter, Phys. Rev. D22, 738 (1980)

  14. [14]

    Pisano and V

    F. Pisano and V. Pleitez, Phys. Rev. D46, 410 (1992), arXiv:hep-ph/9206242

  15. [15]

    Search for doubly-charged Higgs bosons in like-sign dilepton final states at sqrt(s) = 7 TeV with the ATLAS detector

    G. Aadet al.(ATLAS), Eur. Phys. J. C72, 2244 (2012), arXiv:1210.5070 [hep-ex]

  16. [16]

    A search for a doubly-charged Higgs boson in pp collisions at sqrt(s) = 7 TeV

    S. Chatrchyanet al.(CMS), Eur. Phys. J. C72, 2189 (2012), arXiv:1207.2666 [hep-ex]

  17. [17]

    Aaboudet al.(ATLAS), Eur

    M. Aaboudet al.(ATLAS), Eur. Phys. J. C78, 199 (2018), arXiv:1710.09748 [hep-ex]

  18. [18]

    Aadet al.(ATLAS), JHEP06, 146, arXiv:2101.11961 [hep-ex]

    G. Aadet al.(ATLAS), JHEP06, 146, arXiv:2101.11961 [hep-ex]

  19. [19]

    Exploring the Doubly Charged Higgs of the Left-Right Symmetric Model using Vector Boson Fusion-like Events at the LHC

    B. Dutta, R. Eusebi, Y. Gao, T. Ghosh, and T. Kamon, Phys. Rev. D90, 055015 (2014), arXiv:1404.0685 [hep- ph]

  20. [20]

    K. S. Babu and S. Jana, Phys. Rev. D95, 055020 (2017), arXiv:1612.09224 [hep-ph]

  21. [21]

    Exclusive doubly charged Higgs boson pair production in $pp$ collisions at the LHC

    L. Duarte, V. P. Goncalves, D. E. Martins, T. B. de Melo, and F. S. Queiroz, Phys. Rev. D107, 035010 (2023), arXiv:2208.07599 [hep-ph]

  22. [22]

    Ashanujjaman, K

    S. Ashanujjaman, K. Ghosh, and R. Sahu, Phys. Rev. D 107, 015018 (2023), arXiv:2211.00632 [hep-ph]

  23. [23]

    Ashanujjaman and S

    S. Ashanujjaman and S. P. Maharathy, Phys. Rev. D107, 115026 (2023), arXiv:2305.06889 [hep-ph]

  24. [24]

    P. S. Bhupal Dev and Y. Zhang, JHEP10, 199, arXiv:1808.00943 [hep-ph]

  25. [25]

    Ashanujjaman and K

    S. Ashanujjaman and K. Ghosh, JHEP03, 195, arXiv:2108.10952 [hep-ph]

  26. [26]

    Type-II see-saw: searching the LHC elusive low-mass triplet-like Higgses at $e^-e^+$ colliders

    S. Ashanujjaman, K. Ghosh, and K. Huitu, Phys. Rev. D106, 075028 (2022), arXiv:2205.14983 [hep-ph]

  27. [27]

    Aadet al.(ATLAS), Eur

    G. Aadet al.(ATLAS), Eur. Phys. J. C83, 605 (2023), arXiv:2211.07505 [hep-ex]

  28. [28]

    8 ch/Atlas/GROUPS/PHYSICS/PAPERS/EXOT-2018-34/ index.php

    ATLAS Collaboration, Auxiliary material for atlas search exot-2018-34 (2022),https://atlas.web.cern. 8 ch/Atlas/GROUPS/PHYSICS/PAPERS/EXOT-2018-34/ index.php

  29. [29]

    P. D. Bolton, J. Kriewald, M. Nemevˇ sek, F. Nesti, and J. C. Vasquez, Phys. Rev. D111, 035016 (2025), arXiv:2408.00833 [hep-ph]

  30. [30]

    On the Robustness of type-II Seesaw Collider Searches

    C. Englert, M. Mitra, W. Naskar, and S. Saha, (2026), arXiv:2603.09244 [hep-ph]

  31. [31]

    R. L. Workmanet al.(Particle Data Group), PTEP 2022, 083C01 (2022)

  32. [32]

    Aadet al.(ATLAS), JINST14(12), P12006, arXiv:1908.00005 [hep-ex]

    G. Aadet al.(ATLAS), JINST14(12), P12006, arXiv:1908.00005 [hep-ex]

  33. [33]

    Identification and energy calibration of hadronic tau lepton decays at the LHC

    M. Flechl (ATLAS, CMS), in5th Large Hadron Collider Physics Conference(2017) arXiv:1709.01351 [hep-ex]

  34. [34]

    Alwall, R

    J. Alwall, R. Frederix, S. Frixione, V. Hirschi, F. Maltoni, O. Mattelaer, H. S. Shao, T. Stelzer, P. Torrielli, and M. Zaro, JHEP07, 079, arXiv:1405.0301 [hep-ph]

  35. [35]

    B. Fuks, M. Nemevˇ sek, and R. Ruiz, Phys. Rev. D101, 075022 (2020), arXiv:1912.08975 [hep-ph]

  36. [36]
  37. [37]

    Sj¨ ostrand, S

    T. Sj¨ ostrand, S. Ask, J. R. Christiansen, R. Corke, N. De- sai, P. Ilten, S. Mrenna, S. Prestel, C. O. Rasmussen, and P. Z. Skands, Comput. Phys. Commun.191, 159 (2015), arXiv:1410.3012 [hep-ph]

  38. [38]

    de Favereau, C

    J. de Favereau, C. Delaere, P. Demin, A. Giammanco, V. Lema ˆ ıtre, A. Mertens, and M. Selvaggi (DELPHES 3), JHEP02, 057, arXiv:1307.6346 [hep-ex]

  39. [39]

    Cacciari, G

    M. Cacciari, G. P. Salam, and G. Soyez, JHEP04, 063, arXiv:0802.1189 [hep-ph]

  40. [40]

    Aaboudet al.(ATLAS), Eur

    M. Aaboudet al.(ATLAS), Eur. Phys. J. C79, 639 (2019), arXiv:1902.04655 [physics.ins-det]

  41. [41]

    Aadet al.(ATLAS), Eur

    G. Aadet al.(ATLAS), Eur. Phys. J. C81, 578 (2021), arXiv:2012.00578 [hep-ex]

  42. [42]

    Aadet al.(ATLAS), Eur

    G. Aadet al.(ATLAS), Eur. Phys. J. C76, 292 (2016), arXiv:1603.05598 [hep-ex]

  43. [43]

    A. L. Read, J. Phys. G28, 2693 (2002)

  44. [44]

    pyhf: pure-Python implementation of HistFactory with tensors and automatic differentiation

    M. Feickert, L. Heinrich, and G. Stark, PoS ICHEP2022, 245 (2022), arXiv:2211.15838 [hep-ex]

  45. [45]

    Heinrich, M

    L. Heinrich, M. Feickert, G. Stark, and K. Cranmer, J. Open Source Softw.6, 2823 (2021)