pith. sign in

arxiv: 2507.02032 · v2 · pith:ARI32V74new · submitted 2025-07-02 · ✦ hep-ph · hep-ex· physics.data-an· stat.ML

Neural simulation-based inference of the Higgs trilinear self-coupling via off-shell Higgs production

Pith reviewed 2026-05-21 23:26 UTC · model grok-4.3

classification ✦ hep-ph hep-exphysics.data-anstat.ML
keywords Higgs trilinear couplingoff-shell Higgs productionsimulation-based inferenceSMEFTneural networksquantum interferenceHL-LHCbackground estimation
0
0 comments X

The pith

A hybrid neural simulation-based inference method extracts the Higgs trilinear self-coupling from off-shell production data with sensitivity near the theoretical optimum.

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

This paper develops a hybrid neural simulation-based inference technique to extract the Higgs trilinear self-coupling by analyzing off-shell Higgs production in proton-proton collisions. The approach builds a likelihood that includes Standard Model effective field theory modifications to the signal, all relevant background processes, and quantum interference effects between them. It combines the efficiency of matrix-element methods for handling new physics with the practicality of classification techniques for background modeling. The resulting estimator reaches performance close to the best possible limit set by the data, yielding projected constraints for the high-luminosity LHC while also addressing other operators that influence the same final states. A reader would care because this self-coupling determines the shape of the Higgs potential, and off-shell production offers an independent experimental handle on it.

Core claim

We design a hybrid neural simulation-based inference (NSBI) approach to construct a likelihood of the Higgs signal incorporating modifications from the Standard Model effective field theory (SMEFT), relevant background processes, and quantum interference effects. It leverages the training efficiency of matrix-element-enhanced techniques, which are vital for robust SMEFT applications, while also incorporating the practical advantages of classification-based methods for effective background estimates. We demonstrate that our NSBI approach achieves sensitivity close to the theoretical optimum and provide expected constraints for the high-luminosity upgrade of the Large Hadron Collider. While we

What carries the argument

The hybrid NSBI likelihood estimator that merges matrix-element-enhanced training for signal and interference modeling with classification-based background estimation to produce an unbiased estimator over the full phase space.

If this is right

  • Expected constraints on the Higgs trilinear self-coupling can be obtained at the high-luminosity LHC that are competitive with on-shell channels.
  • The same framework simultaneously constrains additional SMEFT operators that modify off-shell Higgs production.
  • The estimator remains unbiased when quantum interference and new-physics effects are present in the data.
  • Performance approaches the theoretical optimum set by the information content of the selected final states.

Where Pith is reading between the lines

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

  • The computational efficiency of the hybrid training could allow the inclusion of additional final states or higher-order corrections without prohibitive cost.
  • Combining the off-shell results with on-shell double-Higgs measurements in a joint likelihood would reduce correlations between the trilinear coupling and other operators.
  • Validation on real collision data would require testing how well the method handles detector-level effects that are only approximated in the current simulation.

Load-bearing premise

The hybrid training procedure combining matrix-element-enhanced techniques with classification-based background estimates produces an unbiased likelihood estimator that correctly incorporates quantum interference and SMEFT modifications across the full phase space.

What would settle it

A direct comparison on the same set of simulated events between the sensitivity achieved by the trained NSBI model and the sensitivity of an idealized matrix-element likelihood analysis that uses the true parton-level information would show whether the method truly reaches near-optimal performance.

read the original abstract

One of the forthcoming major challenges in particle physics is the experimental determination of the Higgs trilinear self-coupling. While efforts have largely focused on on-shell double- and single-Higgs production in proton-proton collisions, off-shell Higgs production has also been proposed as a valuable complementary probe. In this article, we design a hybrid neural simulation-based inference (NSBI) approach to construct a likelihood of the Higgs signal incorporating modifications from the Standard Model effective field theory (SMEFT), relevant background processes, and quantum interference effects. It leverages the training efficiency of matrix-element-enhanced techniques, which are vital for robust SMEFT applications, while also incorporating the practical advantages of classification-based methods for effective background estimates. We demonstrate that our NSBI approach achieves sensitivity close to the theoretical optimum and provide expected constraints for the high-luminosity upgrade of the Large Hadron Collider. While we primarily concentrate on the Higgs trilinear self-coupling, we also consider constraints on other SMEFT operators that affect off-shell Higgs production.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

1 major / 2 minor

Summary. The paper introduces a hybrid neural simulation-based inference (NSBI) method for constraining the Higgs trilinear self-coupling using off-shell Higgs production at the LHC. The approach combines matrix-element-enhanced techniques for SMEFT efficiency with classification-based background modeling to construct a likelihood that incorporates signal modifications, backgrounds, and quantum interference effects. The authors claim this yields sensitivity close to the theoretical optimum and provide projected constraints for the HL-LHC, while also considering other SMEFT operators affecting off-shell production.

Significance. If validated, the hybrid NSBI framework could provide a useful complementary probe of the Higgs potential via off-shell kinematics, where interference effects are prominent. The method's emphasis on handling SMEFT modifications efficiently across phase space addresses a practical challenge in simulation-based inference for high-dimensional LHC analyses.

major comments (1)
  1. [Section describing the hybrid NSBI method and results on sensitivity] The central claim that the NSBI approach achieves sensitivity close to the theoretical optimum rests on the hybrid training procedure producing an unbiased likelihood estimator that correctly incorporates SMEFT modifications and interference over the full off-shell phase space. Explicit closure tests or similar validation demonstrating that residual mismatches between the matrix-element and classification components do not bias the likelihood ratio are required; without them the reported sensitivity and HL-LHC projections cannot be considered robust.
minor comments (2)
  1. [Results section] Clarify the precise definition of 'theoretical optimum' used for the sensitivity comparison and specify how it is computed in the presence of backgrounds and interference.
  2. [Figures presenting HL-LHC projections] Ensure all figures showing expected constraints include statistical and systematic uncertainty bands for direct comparison with the claimed near-optimal performance.

Simulated Author's Rebuttal

1 responses · 0 unresolved

We thank the referee for their careful reading and constructive feedback on our manuscript. The single major comment raises an important point about validation of the hybrid NSBI likelihood estimator. We address this below and have revised the manuscript accordingly to strengthen the robustness of our claims.

read point-by-point responses
  1. Referee: [Section describing the hybrid NSBI method and results on sensitivity] The central claim that the NSBI approach achieves sensitivity close to the theoretical optimum rests on the hybrid training procedure producing an unbiased likelihood estimator that correctly incorporates SMEFT modifications and interference over the full off-shell phase space. Explicit closure tests or similar validation demonstrating that residual mismatches between the matrix-element and classification components do not bias the likelihood ratio are required; without them the reported sensitivity and HL-LHC projections cannot be considered robust.

    Authors: We agree that explicit validation is essential to substantiate the unbiased character of the hybrid likelihood estimator. In the revised manuscript we have added a dedicated subsection (now Section 4.3) presenting closure tests. These tests compare the hybrid NSBI likelihood ratios against exact matrix-element ratios in controlled toy scenarios that isolate SMEFT modifications and interference effects across the off-shell phase space. The results show that any residual mismatches between the matrix-element-enhanced and classification-based components remain below the percent level and do not introduce measurable bias in the extracted constraints. We have also updated the discussion of the HL-LHC projections to reference these tests explicitly, thereby reinforcing the robustness of the reported sensitivity. revision: yes

Circularity Check

0 steps flagged

No circularity: NSBI method derives sensitivity from simulation-based likelihood construction

full rationale

The paper presents a hybrid neural simulation-based inference (NSBI) framework that combines matrix-element-enhanced training for SMEFT effects with classification-based background modeling to construct a likelihood estimator over the off-shell phase space. The reported sensitivity to the Higgs trilinear coupling and HL-LHC projections are obtained by applying this estimator to simulated data under the assumed model, without evidence that any central result reduces by construction to a fit on the target observables or to a self-citation chain. The derivation chain remains self-contained against external benchmarks because the unbiasedness claim rests on the explicit hybrid training procedure rather than re-labeling inputs as outputs. No self-definitional, fitted-input-as-prediction, or ansatz-smuggled steps are identifiable from the method description.

Axiom & Free-Parameter Ledger

0 free parameters · 1 axioms · 0 invented entities

Abstract provides insufficient detail to enumerate specific free parameters or invented entities; the approach relies on standard SMEFT assumptions and neural-network training on simulated events whose hyperparameters are not listed.

axioms (1)
  • domain assumption SMEFT is a valid effective description of possible new physics affecting off-shell Higgs production
    Invoked when incorporating modifications from SMEFT operators into the signal likelihood.

pith-pipeline@v0.9.0 · 5722 in / 1301 out tokens · 40625 ms · 2026-05-21T23:26:31.735949+00:00 · methodology

discussion (0)

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

Lean theorems connected to this paper

Citations machine-checked in the Pith Canon. Every link opens the source theorem in the public Lean library.

What do these tags mean?
matches
The paper's claim is directly supported by a theorem in the formal canon.
supports
The theorem supports part of the paper's argument, but the paper may add assumptions or extra steps.
extends
The paper goes beyond the formal theorem; the theorem is a base layer rather than the whole result.
uses
The paper appears to rely on the theorem as machinery.
contradicts
The paper's claim conflicts with a theorem or certificate in the canon.
unclear
Pith found a possible connection, but the passage is too broad, indirect, or ambiguous to say the theorem truly supports the claim.

Reference graph

Works this paper leans on

64 extracted references · 64 canonical work pages · 30 internal anchors

  1. [2]

    JHEP 12, 148 (2022) https://doi.org/ 10.1007/JHEP12(2022)148 arXiv:2209.00666 [hep-ph]

    Durieux, G., McCullough, M., Salvioni, E.: Charting the Higgs self-coupling bound- aries. JHEP 12, 148 (2022) https://doi.org/ 10.1007/JHEP12(2022)148 arXiv:2209.00666 [hep-ph]. [Erratum: JHEP02, 165 (2023)]

  2. [3]

    Indirect probes of the trilinear Higgs coupling: $gg \to h$ and $h \to \gamma \gamma$

    Gorbahn, M., Haisch, U.: Indirect probes of the trilinear Higgs coupling:gg → h and h → γγ. JHEP 10, 094 (2016) https://doi.org/ 10.1007/JHEP10(2016)094 arXiv:1607.03773 [hep-ph]

  3. [4]

    JHEP 12, 080 (2016) https://doi.org/

    Degrassi, G., Giardino, P.P., Maltoni, F., Pagani, D.: Probing the Higgs self cou- pling via single Higgs production at the LHC. JHEP 12, 080 (2016) https://doi.org/

  4. [5]

    0 MC estimate pq ¯q(x) pq ¯q(x) + pgg(x) pp → ZZ → 2ℓ2ν √s = 14 TeV MC NSBI

  5. [6]

    0 0. 2 0. 4 0. 6 0. 8 1. 0 NSBI estimate ˆs(x) −0. 05

  6. [7]

    Probing the Higgs self coupling via single Higgs production at the LHC

    05 Residual 1 Fig. 14: Calibration closure of the NN-estimated probability ratio, similar to Figure 9, but for the pp → ZZ → 2ℓ2ν channel. The results displayed include events from thegg → (h∗ →) ZZ → 2ℓ2ν, q¯q → ZZ → 2ℓ2ν, and q¯q → W +W − → 2ℓ2ν channels. 10.1007/JHEP12(2016)080 arXiv:1607.04251 [hep-ph]

  7. [8]

    Constraints on the trilinear Higgs coupling from vector boson fusion and associated Higgs production at the LHC

    Bizoń, W., Gorbahn, M., Haisch, U., Zan- derighi, G.: Constraints on the trilinear Higgs coupling from vector boson fusion and associated Higgs production at the LHC. JHEP 07, 083 (2017) https://doi.org/ 10.1007/JHEP07(2017)083 arXiv:1610.05771 [hep-ph]

  8. [9]

    A global view on the Higgs self-coupling

    Di Vita, S., Grojean, C., Panico, G., Riembau, M., Vantalon, T.: A global view on the Higgs self-coupling. JHEP 09, 069 (2017) https://doi.org/10.1007/ JHEP09(2017)069 arXiv:1704.01953 [hep- ph]

  9. [11]

    Two-loop amplitudes for Higgs plus jet production involving a modified trilinear Higgs coupling

    Gorbahn, M., Haisch, U.: Two-loop ampli- tudes for Higgs plus jet production involv- ing a modified trilinear Higgs coupling. JHEP 04, 062 (2019) https://doi.org/ 10.1007/JHEP04(2019)062 arXiv:1902.05480 [hep-ph]

  10. [14]

    JHEP 10, 236 (2024) https://doi.org/ 10.1007/JHEP10(2024)236 arXiv:2408.13186 [hep-ph]

    Haisch, U., Niggetiedt, M.: Exact two- loop amplitudes for Higgs plus jet pro- duction with a cubic Higgs self-coupling. JHEP 10, 236 (2024) https://doi.org/ 10.1007/JHEP10(2024)236 arXiv:2408.13186 [hep-ph]

  11. [15]

    : Constraints on the Higgs boson self-coupling from single- and double- Higgs production with the ATLAS detector using pp collisions at √s = 13 TeV

    Aad, G., et al. : Constraints on the Higgs boson self-coupling from single- and double- Higgs production with the ATLAS detector using pp collisions at √s = 13 TeV. Phys. Lett. B 843, 137745 (2023) https: //doi.org/10.1016/j.physletb.2023.137745 arXiv:2211.01216 [hep-ex]

  12. [16]

    : Constraints on the Higgs boson self-coupling from the combination of single and double Higgs boson production in proton- proton collisions at √s = 13 TeV

    Hayrapetyan, A., et al. : Constraints on the Higgs boson self-coupling from the combination of single and double Higgs boson production in proton- proton collisions at √s = 13 TeV. Phys. Lett. B 861, 139210 (2025) https: //doi.org/10.1016/j.physletb.2024.139210 arXiv:2407.13554 [hep-ex]

  13. [17]

    David, A.,Denner, A.,Duehrssen, M.,Grazz- ini, M., Grojean, C., Passarino, G., Schu- macher, M., Spira, M., Weiglein, G., Zanetti, M.: LHC HXSWG interim recommendations to explore the coupling structure of a Higgs- like particle (2012) arXiv:1209.0040 [hep-ph]

  14. [18]

    Handbook of LHC Higgs Cross Sections: 3. Higgs Properties

    Andersen, J.R., et al .: Handbook of LHC Higgs Cross Sections: 3. Higgs Prop- erties (2013) https://doi.org/10.5170/ CERN-2013-004 arXiv:1307.1347 [hep-ph]

  15. [19]

    Handbook of LHC Higgs Cross Sections: 4. Deciphering the Nature of the Higgs Sector

    Florian, D., et al .: Handbook of LHC Higgs Cross Sections: 4. Deciphering the Nature of the Higgs Sector 2/2017 (2016) https://doi.org/10.23731/CYRM-2017-002 arXiv:1610.07922 [hep-ph]

  16. [20]

    Feruglio, F.: The Chiral approach to the elec- troweak interactions. Int. J. Mod. Phys. A8, 4937–4972 (1993) https://doi.org/10.1142/ S0217751X93001946 arXiv:hep-ph/9301281

  17. [21]

    Grinstein, B., Trott, M.: A Higgs-Higgs bound state due to new physics at a TeV. Phys. Rev. D 76, 073002 (2007) https: //doi.org/10.1103/PhysRevD.76.073002 arXiv:0704.1505 [hep-ph]

  18. [22]

    Effective Theory of a Dynamically Broken Electroweak Standard Model at NLO

    Buchalla, G., Catà, O.: Effective Theory of a Dynamically Broken Electroweak Stan- dard Model at NLO. JHEP 07, 101 (2012) https://doi.org/10.1007/JHEP07(2012)101 arXiv:1203.6510 [hep-ph]

  19. [23]

    Alonso, R., Gavela, M.B., Merlo, L., Rigolin, S.,Yepes,J.:TheEffectiveChiralLagrangian for a Light Dynamical ”Higgs Particle”. Phys. Lett. B 722, 330–335 (2013) https: //doi.org/10.1016/j.physletb.2013.04.037 arXiv:1212.3305 [hep-ph]. [Erratum: Phys. Lett. B 726, 926 (2013)]

  20. [24]

    Buchalla, G., Catà, O., Krause, C.: Complete Electroweak Chiral Lagrangian with a Light Higgs at NLO. Nucl. Phys. B880, 552–573 (2014) https://doi.org/10.1016/j.nuclphysb. 2014.01.018 arXiv:1307.5017 [hep-ph]. [Erra- tum: Nucl. Phys. B913, 475–478 (2016)]

  21. [25]

    Buchmüller, W., Wyler, D.: Effective Lagrangian Analysis of New Interac- tions and Flavor Conservation. Nucl. Phys. B 268, 621–653 (1986) https: //doi.org/10.1016/0550-3213(86)90262-2 21

  22. [26]

    Dimension-Six Terms in the Standard Model Lagrangian

    Grzadkowski, B., Iskrzynski, M., Misiak, M., Rosiek, J.: Dimension-Six Terms in the Standard Model Lagrangian. JHEP 10, 085 (2010) https://doi.org/10.1007/ JHEP10(2010)085 arXiv:1008.4884 [hep-ph]

  23. [28]

    Isidori, G., Wilsch, F., Wyler, D.: The stan- dard model effective field theory at work. Rev. Mod. Phys.96(1), 015006 (2024) https: //doi.org/10.1103/RevModPhys.96.015006 arXiv:2303.16922 [hep-ph]

  24. [29]

    JHEP02, 030 (2022) https://doi.org/10.1007/JHEP02(2022)030 arXiv:2111.12589 [hep-ph]

    Haisch, U., Koole, G.: Off-shell Higgs production at the LHC as a probe of the tri- linear Higgs coupling. JHEP02, 030 (2022) https://doi.org/10.1007/JHEP02(2022)030 arXiv:2111.12589 [hep-ph]

  25. [30]

    Gonçalves, D., Han, T., Mukhopadhyay, S.: Off-Shell Higgs Probe of Naturalness. Phys. Rev. Lett. 120(11), 111801 (2018) https: //doi.org/10.1103/PhysRevLett.120.111801 arXiv:1710.02149 [hep-ph]. [Erratum: Phys. Rev. Lett. 121, 079902 (2018)]

  26. [31]

    Gonçalves, D., Han, T., Mukhopadhyay, S.: Higgs Couplings at High Scales. Phys. Rev. D 98(1), 015023 (2018) https://doi.org/10. 1103/PhysRevD.98.015023 arXiv:1803.09751 [hep-ph]

  27. [32]

    JHEP 04, 166 (2022) https://doi.org/ 10.1007/JHEP04(2022)166 arXiv:2201.09711 [hep-ph]

    Haisch, U., Koole, G.: Probing Higgs portals with matrix-element based kine- matic discriminants in ZZ → 4ℓ produc- tion. JHEP 04, 166 (2022) https://doi.org/ 10.1007/JHEP04(2022)166 arXiv:2201.09711 [hep-ph]

  28. [33]

    SciPost Phys

    Haisch, U., Ruhdorfer, M., Schmid, K., Weiler, A.: Quantum collider probes of the fermionic Higgs portal. SciPost Phys. 16(4), 112 (2024) https://doi.org/10.21468/ SciPostPhys.16.4.112 arXiv:2311.03995 [hep- ph]

  29. [34]

    JHEP 10, 027 (2023) https://doi.org/10.1007/JHEP10(2023)027 arXiv:2304.09772 [hep-ph]

    Balzani, E., Gröber, R., Vitti, M.: Light- quark Yukawa couplings from off-shell Higgs production. JHEP 10, 027 (2023) https://doi.org/10.1007/JHEP10(2023)027 arXiv:2304.09772 [hep-ph]

  30. [36]

    Brehmer, J., Cranmer, K., Louppe, G., Pavez, J.: A Guide to Constraining Effec- tive Field Theories with Machine Learning. Phys. Rev. D 98(5), 052004 (2018) https: //doi.org/10.1103/PhysRevD.98.052004 arXiv:1805.00020 [hep-ph]

  31. [37]

    Brehmer, J., Kling, F., Espejo, I., Cranmer, K.: MadMiner: Machine learning-based infer- ence for particle physics. Comput. Softw. Big Sci. 4(1), 3 (2020) https://doi.org/10.1007/ s41781-020-0035-2 arXiv:1907.10621 [hep- ph]

  32. [38]

    Cranmer, K., Brehmer, J., Louppe, G.: The frontier of simulation-based inference. Proc. Nat. Acad. Sci. 117(48), 30055– 30062 (2020) https://doi.org/10.1073/pnas. 1912789117 arXiv:1911.01429 [stat.ML]

  33. [39]

    https://hal

    Ghosh, A.: Measuring quantum interference in the off-shell Higgs to four leptons process with Machine Learning (2020). https://hal. archives-ouvertes.fr/hal-02971995

  34. [40]

    Aad, G., et al. : Measurement of off-shell Higgs boson production in theH ∗ → ZZ → 4ℓ decay channel using a neural simulation- based inference technique in 13 TeVpp colli- sions with the ATLAS detector. Rept. Prog. Phys. 88(5), 057803 (2025) https://doi.org/ 10.1088/1361-6633/adcd9a arXiv:2412.01548 [hep-ex]

  35. [41]

    : An implementation of neural simulation-based inference for parameter estimation in ATLAS

    Aad, G., et al. : An implementation of neural simulation-based inference for parameter estimation in ATLAS. Rept. Prog. Phys. 88(6), 067801 (2025) https://doi.org/10.1088/1361-6633/add370 22 arXiv:2412.01600 [physics.data-an]

  36. [42]

    Bounding the Higgs width at the LHC using full analytic results for gg -> 2e 2\mu

    Campbell, J.M., Ellis, R.K., Williams, C.: Bounding the Higgs Width at the LHC Using Full Analytic Results for gg → e−e+µ−µ+. JHEP 04, 060 (2014) https://doi.org/ 10.1007/JHEP04(2014)060 arXiv:1311.3589 [hep-ph]

  37. [43]

    : Higgs boson production cross-section measurements and their EFT interpretation in the 4ℓ decay channel at√s = 13 TeV with the ATLAS detector

    Aad, G., et al. : Higgs boson production cross-section measurements and their EFT interpretation in the 4ℓ decay channel at√s = 13 TeV with the ATLAS detector. Eur. Phys. J. C 80(10), 957 (2020) https: //doi.org/10.1140/epjc/s10052-020-8227-9 arXiv:2004.03447 [hep-ex]. [Erratum: Eur. Phys. J. C 81, 29 (2021), Erratum: Eur. Phys. J. C81, 398 (2021)]

  38. [44]

    https://cds.cern

    ATLAS: Measurements and interpretation of the Higgs boson differential and production mode cross sections in theH → ZZ ∗ → 4ℓ channel at √s = 13.6 TeV with the ATLAS detector, Geneva (2025). https://cds.cern. ch/record/2929042

  39. [45]

    JHEP 01, 192 (2024) https: //doi.org/10.1007/JHEP01(2024)192 arXiv:2311.06107 [hep-ph]

    Gauld, R., Haisch, U., Schnell, L.: SMEFT at NNLO+PS: V h produc- tion. JHEP 01, 192 (2024) https: //doi.org/10.1007/JHEP01(2024)192 arXiv:2311.06107 [hep-ph]

  40. [46]

    Probing the scalar potential via double Higgs boson production at hadron colliders

    Borowka, S., Duhr, C., Maltoni, F., Pagani, D., Shivaji, A., Zhao, X.: Probing the scalar potential via double Higgs boson production at hadron colliders. JHEP 04, 016 (2019) https://doi.org/10.1007/JHEP04(2019)016 arXiv:1811.12366 [hep-ph]

  41. [47]

    Boughezal, R., Campbell, J.M., Ellis, R.K., Focke, C., Giele, W., Liu, X., Petriello, F., Williams, C.: Color sin- glet production at NNLO in MCFM. Eur. Phys. J. C 77(1), 7 (2017) https: //doi.org/10.1140/epjc/s10052-016-4558-y arXiv:1605.08011 [hep-ph]

  42. [48]

    Gritsan, A.V., Roskes, J., Sarica, U., Schulze, M., Xiao, M., Zhou, Y.: New fea- tures in the JHU generator framework: constraining Higgs boson properties from on-shell and off-shell production. Phys. Rev. D 102(5), 056022 (2020) https: //doi.org/10.1103/PhysRevD.102.056022 arXiv:2002.09888 [hep-ph]

  43. [49]

    Navas, S., et al.: Review of particle physics. Phys. Rev. D 110(3), 030001 (2024) https: //doi.org/10.1103/PhysRevD.110.030001

  44. [50]

    Denner, A., Dittmaier, S., Roth, M., Wackeroth, D.: Electroweak radia- tive corrections to e+e− → W W → 4 fermions in double pole approxima- tion: The RACOONWW approach. Nucl. Phys. B 587, 67–117 (2000) https: //doi.org/10.1016/S0550-3213(00)00511-3 arXiv:hep-ph/0006307

  45. [51]

    Ball, R.D., et al.: The path to proton struc- ture at 1% accuracy. Eur. Phys. J. C82(5), 428 (2022) https://doi.org/10.1140/epjc/ s10052-022-10328-7 arXiv:2109.02653 [hep- ph]

  46. [52]

    JHEP 01, 072 (2022) https://doi.org/ 10.1007/JHEP01(2022)072 arXiv:2108.05337 [hep-ph]

    Buonocore, L., Koole, G., Lombardi, D., Rot- toli, L., Wiesemann, M., Zanderighi, G.:ZZ production at nNNLO+PS with MiNNLOPS. JHEP 01, 072 (2022) https://doi.org/ 10.1007/JHEP01(2022)072 arXiv:2108.05337 [hep-ph]

  47. [53]

    JHEP 03, 070 (2019) https://doi.org/10.1007/JHEP03(2019)070 arXiv:1811.09593 [hep-ph]

    Grazzini, M., Kallweit, S., Wiesemann, M., Yook, J.Y.: ZZ production at the LHC: NLO QCD corrections to the loop-induced gluon fusion channel. JHEP 03, 070 (2019) https://doi.org/10.1007/JHEP03(2019)070 arXiv:1811.09593 [hep-ph]

  48. [56]

    Off-Shell Higgs Coupling Measurements in BSM scenarios

    Englert, C., Soreq, Y., Spannowsky, M.: Off-Shell Higgs Coupling Measurements in BSM scenarios. JHEP 05, 145 (2015) https://doi.org/10.1007/JHEP05(2015)145 arXiv:1410.5440 [hep-ph]

  49. [57]

    Cambridge University Press, Cambridge (2012)

    Sugiyama, M., Suzuki, T., Kanamori, T.: Density Ratio Estimation in Machine Learn- ing. Cambridge University Press, Cambridge (2012)

  50. [58]

    Cranmer, K., Pavez, J., Louppe, G.: Approximating Likelihood Ratios with Cal- ibrated Discriminative Classifiers (2015) arXiv:1506.02169 [stat.AP]

  51. [59]

    Paszke, A., et al.: PyTorch: An Impera- tive Style, High-Performance Deep Learning Library (2019) arXiv:1912.01703 [cs.LG]

  52. [60]

    https://doi.org/ 10.5281/zenodo.7447212

    Falcon, W., et al.: Lightning-AI/lightning- bolts: Minor Patch Release. https://doi.org/ 10.5281/zenodo.7447212

  53. [61]

    Searching for Activation Functions

    Ramachandran, P., Zoph, B., Le, Q.V.: Searching for activation functions. CoRR abs/1710.05941 (2017) 1710.05941

  54. [62]

    Benevedes, S., Thaler, J.: Frequentist Uncer- tainties on Neural Density Ratios with wifi Ensembles (2025) arXiv:2506.00113 [hep-ph]

  55. [63]

    https://cds.cern.ch/record/ 2860128

    ATLAS: Effective field theory interpretation of the measurement of off-shell Higgs boson production from ZZ → 4ℓ and ZZ → 2ℓ2ν decay channels with the ATLAS detector, Geneva (2023). https://cds.cern.ch/record/ 2860128

  56. [64]

    Probing Higgs couplings with high p_T Higgs production

    Azatov, A., Paul, A.: Probing Higgs cou- plings with highpT Higgs production. JHEP 01, 014 (2014) https://doi.org/10.1007/ JHEP01(2014)014 arXiv:1309.5273 [hep-ph]

  57. [65]

    Very boosted Higgs in gluon fusion

    Grojean, C., Salvioni, E., Schlaffer, M., Weiler, A.: Very boosted Higgs in gluon fusion. JHEP05, 022 (2014) https://doi.org/ 10.1007/JHEP05(2014)022 arXiv:1312.3317 [hep-ph]

  58. [66]

    Azatov, A., Grojean, C., Paul, A., Salvioni, E.: Taming the off-shell Higgs boson. Zh. Eksp. Teor. Fiz. 147, 410–425 (2015) https://doi.org/10.1134/S1063776115030140 arXiv:1406.6338 [hep-ph]

  59. [67]

    Mass Effects in the Higgs-Gluon Coupling: Boosted vs Off-Shell Production

    Buschmann, M., Gonçalves, D., Kuttimalai, S., Schönherr, M., Krauss, F., Plehn, T.: Mass Effects in the Higgs-Gluon Coupling: Boosted vs Off-Shell Production. JHEP 02, 038 (2015) https://doi.org/10.1007/ JHEP02(2015)038 arXiv:1410.5806 [hep-ph]

  60. [68]

    https://cds.cern

    ATLAS: HL-LHC prospects for the measure- ment of Higgs boson pair production in the b¯bb¯b final state and combination with the b¯bγγ and b¯bτ +τ − final states at the ATLAS experiment, Geneva (2022). https://cds.cern. ch/record/2841244

  61. [69]

    JHEP 09, 057 (2018) https://doi.org/ 10.1007/JHEP09(2018)057 arXiv:1806.05162 [hep-ph]

    Buchalla, G., Capozi, M., Celis, A., Hein- rich, G., Scyboz, L.: Higgs boson pair production in non-linear Effective Field Theory with full mt-dependence at NLO QCD. JHEP 09, 057 (2018) https://doi.org/ 10.1007/JHEP09(2018)057 arXiv:1806.05162 [hep-ph]. [Erratum: JHEP06, 94 (2025)]

  62. [70]

    Mastandrea, R., Nachman, B., Plehn, T.: Constraining the Higgs potential with neu- ral simulation-based inference for di-Higgs production. Phys. Rev. D 110(5), 056004 (2024) https://doi.org/10.1103/PhysRevD. 110.056004 arXiv:2405.15847 [hep-ph]

  63. [71]

    https://indico.cern.ch/event/ 1355601/overview

    PHYSTAT-SBI 2024 - Simulation Based Inference in Fundamental Physics, Garch- ing (2024). https://indico.cern.ch/event/ 1355601/overview

  64. [72]

    Ellis, J.: TikZ-Feynman: Feynman diagrams with TikZ. Comput. Phys. Commun. 210, 103–123 (2017) https://doi.org/10.1016/j. cpc.2016.08.019 arXiv:1601.05437 [hep-ph] 24