pith. sign in

arxiv: 2512.04153 · v2 · pith:QOCFDN6Bnew · submitted 2025-12-03 · ✦ hep-ph · hep-ex

Data-Driven Predictions for Dark Photon and Millicharged Particle Production

Elizabeth Allison , Nikita Blinov This is my paper

Pith reviewed 2026-05-21 17:59 UTC · model grok-4.3

classification ✦ hep-ph hep-ex
keywords dark photonsmillicharged particlesdata-driven methodsdilepton eventsfixed-target searchesnormalizing flowsnew physics
0
0 comments X

The pith

Measured dilepton events can predict dark photon and millicharged particle production rates and distributions without theoretical models.

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

The paper introduces a framework that forecasts both the rate and kinematic distributions for dark photon (A') and millicharged particle (mCP) production in fixed-target searches by directly using measured dilepton data. It avoids large theoretical uncertainties in hadronic production models by relying instead on the correspondence between emission amplitudes for these new particles and those for off-shell Standard Model photons, which are experimentally accessible. Normalizing flow models learn the relevant distributions from the data to create fast, realistic Monte Carlo generators for signal simulations. A sympathetic reader would care because accurate, model-independent predictions are essential for interpreting results and optimizing experiments that hunt for dark sector physics.

Core claim

By exploiting the close correspondence between the amplitudes for emitting dark photons or millicharged particles and those for producing off-shell Standard Model photons, the authors show that fully differential dilepton data can be used to generate predictions for the rate and kinematics of new physics signals without invoking specific production models.

What carries the argument

The amplitude correspondence between A' or mCP emission and off-shell SM photon production that permits direct reuse of measured dilepton distributions.

If this is right

  • Signal predictions for fixed-target dark sector searches become independent of specific hadronic production models.
  • Normalizing flow models trained on real dilepton data serve as efficient generators for dark photon and mCP kinematics.
  • Experimental design and result interpretation gain reduced theoretical uncertainty.
  • The method applies uniformly to both dark photons and millicharged particles.

Where Pith is reading between the lines

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

  • Existing dilepton datasets from past experiments could be reanalyzed to place new constraints on dark sector parameters.
  • The approach might extend to other particles whose production couples to photons in a similar way, reducing model dependence more broadly.
  • Integration with ongoing detector simulations could yield faster sensitivity estimates for planned runs.

Load-bearing premise

A close correspondence exists between the amplitudes for emission of dark photons or millicharged particles and for off-shell Standard Model photon production.

What would settle it

Direct numerical comparison of the data-driven predictions against those from a complete theoretical Monte Carlo generator on the same simulated dataset, checking for agreement in both rates and kinematic shapes.

Figures

Figures reproduced from arXiv: 2512.04153 by Elizabeth Allison, Nikita Blinov.

Figure 1
Figure 1. Figure 1: FIG. 1: Validation of our mock data set (obtained by reweighing [PITH_FULL_IMAGE:figures/full_fig_p004_1.png] view at source ↗
Figure 3
Figure 3. Figure 3: FIG. 3: Data-driven estimate of mCP ( [PITH_FULL_IMAGE:figures/full_fig_p005_3.png] view at source ↗
Figure 4
Figure 4. Figure 4: FIG. 4: Kullback-Leibler divergence between [PITH_FULL_IMAGE:figures/full_fig_p007_4.png] view at source ↗
Figure 5
Figure 5. Figure 5: FIG. 5: Comparison of the normalized [PITH_FULL_IMAGE:figures/full_fig_p008_5.png] view at source ↗
Figure 6
Figure 6. Figure 6: FIG. 6: Predicted [PITH_FULL_IMAGE:figures/full_fig_p009_6.png] view at source ↗
Figure 7
Figure 7. Figure 7: FIG. 7: Typical training and validation losses as a function of training epoch for the normalizing flow model. [PITH_FULL_IMAGE:figures/full_fig_p010_7.png] view at source ↗
Figure 8
Figure 8. Figure 8: FIG. 8: Kullback-Leibler divergence between the normalizing flow model and the data for different [PITH_FULL_IMAGE:figures/full_fig_p011_8.png] view at source ↗
read the original abstract

Accurate signal predictions are essential for interpreting and optimizing fixed-target searches for new physics. Even in minimal models such as the dark photon ($A'$) or millicharged particles (mCPs), theoretical uncertainties in hadronic production can be substantial. We introduce a data-driven framework that predicts both the rate and kinematic distributions of $A'$ and mCP production directly from measured dilepton events, without relying on specific theoretical production models. This method uses the close correspondence between amplitudes for emission of $A'$ or mCPs, and for off-shell Standard Model photon production, the latter being experimentally measurable in full differential form. We demonstrate that normalizing flow models can learn these distributions from data and serve as a fast, realistic Monte Carlo generator for dark sector signal simulations.

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

2 major / 2 minor

Summary. The manuscript introduces a data-driven framework that uses normalizing flow models trained on measured dilepton events to predict both the rate and kinematic distributions of dark photon (A') and millicharged particle (mCP) production in fixed-target experiments. It relies on an assumed close correspondence between the amplitudes for A'/mCP emission and off-shell Standard Model photon production, allowing direct repurposing of experimental dilepton data without specific theoretical production models for hadronic processes.

Significance. If the amplitude correspondence and flow training hold with controlled systematics, the approach would reduce reliance on uncertain hadronic production models and provide realistic Monte Carlo generators for dark sector signals, strengthening the interpretation of fixed-target searches. The data-driven aspect and emphasis on differential distributions are positive features that could improve sensitivity estimates.

major comments (2)
  1. [§3] §3: The amplitude correspondence is presented as allowing direct use of off-shell dilepton data for on-shell massive A' and mCP cases, but the derivation does not explicitly show the reweighting or resampling procedure for propagator and phase-space differences at finite mass; without this, the learned distributions inherit an unquantified systematic that scales with m_A' and experimental cuts.
  2. [§4] §4: The demonstration that normalizing flows learn the distributions lacks quantitative validation metrics (e.g., Kolmogorov-Smirnov tests, closure tests on injected signals, or comparison to known theoretical benchmarks) and error estimates on the predicted rates, which are required to support the central claim of model-independent predictions.
minor comments (2)
  1. [Abstract] The abstract states the method 'demonstrates' the flows can learn distributions, but this should be supported by explicit figures or tables in the main text showing training convergence and validation.
  2. Notation for the flow model parameters and the precise definition of the 'close correspondence' between amplitudes should be clarified to avoid ambiguity in implementation.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for the constructive and detailed comments on our manuscript. We have addressed each major point below and revised the manuscript accordingly to improve the explicit treatment of the amplitude correspondence and to strengthen the quantitative validation of the normalizing flow approach. These changes enhance the clarity and robustness of our data-driven predictions without altering the core methodology.

read point-by-point responses

  1. Referee: [§3] §3: The amplitude correspondence is presented as allowing direct use of off-shell dilepton data for on-shell massive A' and mCP cases, but the derivation does not explicitly show the reweighting or resampling procedure for propagator and phase-space differences at finite mass; without this, the learned distributions inherit an unquantified systematic that scales with m_A' and experimental cuts.

    Authors: We appreciate the referee's careful reading and agree that an explicit treatment of the finite-mass corrections is necessary for full rigor. In the revised manuscript we have expanded Section 3 with a dedicated derivation that introduces the reweighting factor w(m) = |M_A'(m)|^2 / |M_γ*(q^2 = m^2)|^2 together with the Jacobian for the two-body phase-space mapping between the off-shell photon and the on-shell massive particle. The reweighting is applied after the flow samples the kinematic variables, and we quantify the residual systematic by comparing reweighted and unweighted distributions across the experimental acceptance. For the mass range and cuts relevant to typical fixed-target setups the systematic remains below 5 % and is now reported as a function of m_A' in a new figure. This addition directly addresses the concern while preserving the data-driven character of the method. revision: yes

  2. Referee: [§4] §4: The demonstration that normalizing flows learn the distributions lacks quantitative validation metrics (e.g., Kolmogorov-Smirnov tests, closure tests on injected signals, or comparison to known theoretical benchmarks) and error estimates on the predicted rates, which are required to support the central claim of model-independent predictions.

    Authors: We concur that quantitative validation metrics are essential to substantiate the reliability of the learned distributions. The revised manuscript now includes, in Section 4 and a new Appendix C, Kolmogorov-Smirnov statistics comparing the flow-generated marginals and joint distributions against the input dilepton data. We also present closure tests in which known signal samples (generated from standard theoretical models) are injected into the training set; the flows recover the injected rates and shapes to within the statistical precision of the data. In addition, we provide a comparison of the predicted rates to existing hadronic-production calculations for a benchmark mass point, and we report rate uncertainties obtained from an ensemble of flows trained with different random seeds. These metrics are summarized in a new table and support the model-independent claim within the stated uncertainties. revision: yes

Circularity Check

0 steps flagged

No significant circularity; derivation relies on external amplitude correspondence

full rationale

The paper claims a data-driven prediction of A' and mCP rates and distributions by training normalizing flows on measured dilepton events, justified by an external theoretical correspondence between emission amplitudes and off-shell SM photon production. This correspondence is invoked as an input assumption rather than derived internally, and no equations or steps reduce the output distributions to fitted parameters, self-definitions, or self-citation chains by construction. The framework is therefore self-contained against external benchmarks once the correspondence is granted, with no load-bearing internal reductions that would qualify as circularity.

Axiom & Free-Parameter Ledger

0 free parameters · 1 axioms · 0 invented entities

The central claim depends on the domain assumption of amplitude correspondence between new physics emission and off-shell photon production; no free parameters or invented entities are mentioned in the abstract.

axioms (1)
  • domain assumption Close correspondence exists between amplitudes for A' or mCP emission and off-shell SM photon production.
    This correspondence is invoked to justify using measured dilepton events as a proxy for new physics production.

pith-pipeline@v0.9.0 · 5647 in / 1173 out tokens · 67562 ms · 2026-05-21T17:59:29.831536+00:00 · methodology

discussion (0)

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

Forward citations

Cited by 1 Pith paper

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score.

  1. A new approach to dark photon

    hep-ph 2026-04 unverdicted novelty 5.0

    Dark photon and hypercharge arise from two U(1) groups related by broken mirror symmetry that suppresses their kinetic mixing at one loop.

Reference graph

Works this paper leans on

99 extracted references · 99 canonical work pages · cited by 1 Pith paper · 27 internal anchors

  1. [1]

    The resulting yield per proton on target is shown in Fig

    Dark Photon We compute the totalA ′ cross-section by binning our mock data inm γ and evaluatingdσ γ/dm2 γ in each bin using the weights described above. The resulting yield per proton on target is shown in Fig. 2. Thus a yield prediction can be made in situ, using the detector and target configuration specific to a given experiment. Alongside our (mock) d...

    work page

  2. [2]

    Millicharged Particles We obtain the mCP yield by summing the aforemen- tioned weights for events withm γ >2m χ, multiplied by the prefactor in Eq. (8). The result is shown in Fig. 3. Here we note that experimental selections (such as the focus on dimuon final states, or cuts onm µµ) can ex- clude important production mechanisms, so the (mock) data-driven...

    work page

  3. [3]

    good” value of this metric, we compare this “model vs data

    Dark Photon A measurement of the dilepton invariant mass requires knowledge of the energyq 0 and momentum of the virtual photonq. Thus such a data set should provide a fully dif- ferential estimate ofdσ γ/dm2 γd3q. To make predictions for a dark photon model with a specificm A′, however, we are interested in this distribution with the restriction mγ =m A′...

    work page

  4. [4]

    Normalizing flows can be used to construct a data-driven MC here as well

    Millicharged Particles In contrast to the dark photon case, a wide range mγ ≥2m χ of virtual photon invariant masses contributes to mCP production. Normalizing flows can be used to construct a data-driven MC here as well. Several ap- proaches are possible. The most straightforward, is to train a different flow for eachm χ hypothesis, enforcing mγ ≥2m χ at...

    work page

  5. [5]

    Beachamet al., J

    J. Beacham et al., J. Phys. G47, 010501 (2020), arXiv:1901.09966 [hep-ex]

    work page arXiv 2020

  6. [6]

    Batell, N

    B. Batell, N. Blinov, C. Hearty, and R. McGehee, in Snowmass 2021 (2022) arXiv:2207.06905 [hep-ph]

    work page arXiv 2021

  7. [7]

    Inclusive Dark Photon Search at LHCb

    P. Ilten, Y. Soreq, J. Thaler, M. Williams, and W. Xue, Phys. Rev. Lett.116, 251803 (2016), arXiv:1603.08926 [hep-ph]

    work page internal anchor Pith review Pith/arXiv arXiv 2016

  8. [8]

    J. D. Bjorken, R. Essig, P. Schuster, and N. Toro, Phys. Rev. D80, 075018 (2009), arXiv:0906.0580 [hep-ph]

    work page internal anchor Pith review Pith/arXiv arXiv 2009

  9. [9]

    Resonant production of dark photons in positron beam dump experiments

    E. Nardi, C. D. R. Carvajal, A. Ghoshal, D. Meloni, and M. Raggi, Phys. Rev. D97, 095004 (2018), arXiv:1802.04756 [hep-ph]

    work page internal anchor Pith review Pith/arXiv arXiv 2018

  10. [10]

    Dark photon production through positron annihilation in beam-dump experiments

    L. Marsicano, M. Battaglieri, M. Bondi’, C. D. R. Carvajal, A. Celentano, M. De Napoli, R. De Vita, E. Nardi, M. Raggi, and P. Valente, Phys. Rev. D98, 015031 (2018), arXiv:1802.03794 [hep-ex]

    work page internal anchor Pith review Pith/arXiv arXiv 2018

  11. [11]

    Celentano, L

    A. Celentano, L. Darm´ e, L. Marsicano, and E. Nardi, Phys. Rev. D102, 075026 (2020), arXiv:2006.09419 [hep-ph]

    work page arXiv 2020

  12. [12]

    Kyselov and M

    Y. Kyselov and M. Ovchynnikov, Phys. Rev. D111, 015030 (2025), arXiv:2409.11096 [hep-ph]

    work page arXiv 2025

  13. [13]

    A comprehensive guide to the physics and usage of PYTHIA 8.3

    C. Bierlich et al., SciPost Phys. Codeb.2022, 8 (2022), arXiv:2203.11601 [hep-ph]

    work page internal anchor Pith review Pith/arXiv arXiv 2022

  14. [14]

    Light dark matter in neutrino beams: production modelling and scattering signatures at MiniBooNE, T2K and SHiP

    P. deNiverville, C.-Y. Chen, M. Pospelov, and A. Ritz, Phys. Rev. D95, 035006 (2017), arXiv:1609.01770 [hep-ph]

    work page internal anchor Pith review Pith/arXiv arXiv 2017

  15. [15]

    New Exclusion Limits on Dark Gauge Forces from Proton Bremsstrahlung in Beam-Dump Data

    J. Bl¨ umlein and J. Brunner, Phys. Lett. B731, 320 (2014), arXiv:1311.3870 [hep-ph]

    work page internal anchor Pith review Pith/arXiv arXiv 2014

  16. [16]

    Foroughi-Abari and A

    S. Foroughi-Abari and A. Ritz, Phys. Rev. D105, 095045 (2022), arXiv:2108.05900 [hep-ph]

    work page arXiv 2022

  17. [17]

    Gorbunov and E

    D. Gorbunov and E. Kriukova, JHEP01, 058 (2024), arXiv:2306.15800 [hep-ph]

    work page arXiv 2024

  18. [18]

    Foroughi-Abari, P

    S. Foroughi-Abari, P. Reimitz, and A. Ritz, Phys. Rev. D112, 015030 (2025), arXiv:2409.09123 [hep-ph]

    work page arXiv 2025

  19. [19]

    Kling, P

    F. Kling, P. Reimitz, and A. Ritz, (2025), arXiv:2509.09437 [hep-ph]

    work page arXiv 2025

  20. [20]

    Holdom, Phys

    B. Holdom, Phys. Lett. B166, 196 (1986)

    work page 1986

  21. [21]

    del Aguila, G

    F. del Aguila, G. D. Coughlan, and M. Quiros, Nucl. Phys. B307, 633 (1988), [Erratum: Nucl.Phys.B 312, 751 (1989)]

    work page 1988

  22. [22]

    Lichard, Phys

    P. Lichard, Phys. Rev. D49, 5812 (1994)

    work page 1994

  23. [23]

    Formalism for dilepton production via virtual photon bremsstrahlung in hadronic reactions

    P. Lichard, Phys. Rev. D51, 6017 (1995), arXiv:hep-ph/9812211

    work page internal anchor Pith review Pith/arXiv arXiv 1995

  24. [24]

    Dark Sectors at the Fermilab SeaQuest Experiment

    A. Berlin, S. Gori, P. Schuster, and N. Toro, Phys. Rev. D98, 035011 (2018), arXiv:1804.00661 [hep-ph]

    work page internal anchor Pith review Pith/arXiv arXiv 2018

  25. [25]

    J. M. Berryman, A. de Gouvea, P. J. Fox, B. J. Kayser, K. J. Kelly, and J. L. Raaf, JHEP02, 174 (2020), arXiv:1912.07622 [hep-ph]

    work page arXiv 2020

  26. [26]

    The hadronic light-by-light contribution to the muon's anomalous magnetic moment

    I. Danilkin, C. F. Redmer, and M. Vanderhaeghen, Prog. Part. Nucl. Phys.107, 20 (2019), arXiv:1901.10346 [hep-ph]

    work page internal anchor Pith review Pith/arXiv arXiv 2019

  27. [27]

    Exploring Portals to a Hidden Sector Through Fixed Targets

    B. Batell, M. Pospelov, and A. Ritz, Phys. Rev. D80, 095024 (2009), arXiv:0906.5614 [hep-ph]

    work page internal anchor Pith review Pith/arXiv arXiv 2009

  28. [28]

    Measurement of the $\pi^0$ electromagnetic transition form factor slope

    C. Lazzeroni et al. (NA62), Phys. Lett. B768, 38 (2017), arXiv:1612.08162 [hep-ex]

    work page internal anchor Pith review Pith/arXiv arXiv 2017

  29. [29]

    Precision study of the $\eta\to\mu^+\mu^-\gamma$ and $\omega\to\mu^+\mu^-\pi^0$ electromagnetic transition form-factors and of the $\rho\to\mu^+\mu^-$ line shape in NA60

    R. Arnaldi et al. (NA60), Phys. Lett. B757, 437 (2016), arXiv:1608.07898 [hep-ex]. 12

    work page internal anchor Pith review Pith/arXiv arXiv 2016

  30. [30]

    S. L. Dubovsky, D. S. Gorbunov, and G. I. Rubtsov, JETP Lett.79, 1 (2004), arXiv:hep-ph/0311189

    work page internal anchor Pith review Pith/arXiv arXiv 2004

  31. [31]

    S. D. McDermott, H.-B. Yu, and K. M. Zurek, Phys. Rev. D83, 063509 (2011), arXiv:1011.2907 [hep-ph]

    work page internal anchor Pith review Pith/arXiv arXiv 2011

  32. [32]

    Making dark matter out of light: freeze-in from plasma effects

    C. Dvorkin, T. Lin, and K. Schutz, Phys. Rev. D99, 115009 (2019), [Erratum: Phys.Rev.D 105, 119901 (2022)], arXiv:1902.08623 [hep-ph]

    work page internal anchor Pith review Pith/arXiv arXiv 2019

  33. [33]

    Severely Constraining Dark Matter Interpretations of the 21-cm Anomaly

    A. Berlin, D. Hooper, G. Krnjaic, and S. D. McDermott, Phys. Rev. Lett.121, 011102 (2018), arXiv:1803.02804 [hep-ph]

    work page internal anchor Pith review Pith/arXiv arXiv 2018

  34. [34]

    J. B. Mu˜ noz and A. Loeb, Nature557, 684 (2018), arXiv:1802.10094 [astro-ph.CO]

    work page internal anchor Pith review Pith/arXiv arXiv 2018

  35. [35]

    Holdom, Phys

    B. Holdom, Phys. Lett. B178, 65 (1986)

    work page 1986

  36. [36]

    A. A. Prinz et al., Phys. Rev. Lett.81, 1175 (1998), arXiv:hep-ex/9804008

    work page internal anchor Pith review Pith/arXiv arXiv 1998

  37. [37]

    Millicharged particles in neutrino experiments,

    G. Magill, R. Plestid, M. Pospelov, and Y.-D. Tsai, Phys. Rev. Lett.122, 071801 (2019), arXiv:1806.03310 [hep-ph]

    work page arXiv 2019

  38. [38]

    Acciarri et al

    R. Acciarri et al. (ArgoNeuT), Phys. Rev. Lett.124, 131801 (2020), arXiv:1911.07996 [hep-ex]

    work page arXiv 2020

  39. [39]

    Barak et al

    L. Barak et al. (SENSEI), Phys. Rev. Lett.133, 071801 (2024), arXiv:2305.04964 [hep-ex]

    work page arXiv 2024

  40. [40]

    K. J. Kelly and Y.-D. Tsai, Phys. Rev. D100, 015043 (2019), arXiv:1812.03998 [hep-ph]

    work page arXiv 2019

  41. [41]

    S. N. Gninenko, D. V. Kirpichnikov, and N. V. Krasnikov, Phys. Rev. D100, 035003 (2019), arXiv:1810.06856 [hep-ph]

    work page arXiv 2019

  42. [42]

    Letter of Intent: Search for sub-millicharged particles at J-PARC,

    S. Choi et al., (2020), arXiv:2007.06329 [physics.ins-det]

    work page arXiv 2020

  43. [43]

    Arnaldi et al

    R. Arnaldi et al. (NA60), Eur. Phys. J. C79, 443 (2019), arXiv:2012.14385 [hep-ex]

    work page arXiv 2019

  44. [44]

    Agakichiev et al., Eur

    G. Agakichiev et al., Eur. Phys. J. C4, 231 (1998)

    work page 1998

  45. [45]

    Aberle et al

    O. Aberle et al. (SHiP), BDF/SHiP at the ECN3 high-intensity beam facility, Tech. Rep. (CERN, Geneva, 2022)

    work page 2022

  46. [46]

    Albanese et al

    R. Albanese et al. (SHiP), BDF/SHiP at the ECN3 high-intensity beam facility, Tech. Rep. (CERN, Geneva, 2023)

    work page 2023

  47. [47]

    Apyan et al., in Snowmass 2021 (2022) arXiv:2203.08322 [hep-ex]

    A. Apyan et al., in Snowmass 2021 (2022) arXiv:2203.08322 [hep-ex]

    work page arXiv 2021

  48. [48]

    The active muon shield in the SHiP experiment

    A. Akmete et al. (SHiP), JINST12, P05011 (2017), arXiv:1703.03612 [physics.ins-det]

    work page internal anchor Pith review Pith/arXiv arXiv 2017

  49. [49]

    Navas et al

    S. Navas et al. (Particle Data Group), Phys. Rev. D110, 030001 (2024)

    work page 2024

  50. [50]

    Uras, Low Mass Dimuon Production in p-A Collisions at 400 GeV/cwith the NA60 Detector, Ph.D

    A. Uras, Low Mass Dimuon Production in p-A Collisions at 400 GeV/cwith the NA60 Detector, Ph.D. thesis, Cagliari U. (2011)

    work page 2011

  51. [51]

    Andreassen, P

    A. Andreassen, P. T. Komiske, E. M. Metodiev, B. Nachman, and J. Thaler, Phys. Rev. Lett.124, 182001 (2020), arXiv:1911.09107 [hep-ph]

    work page arXiv 2020

  52. [52]

    Canelli et al., (2025), arXiv:2507.09582 [hep-ph]

    F. Canelli et al., (2025), arXiv:2507.09582 [hep-ph]

    work page arXiv 2025

  53. [53]

    Cowan, Statistical data analysis (1998)

    G. Cowan, Statistical data analysis (1998)

    work page 1998

  54. [54]

    L. Gan, B. Kubis, E. Passemar, and S. Tulin, Phys. Rept.945, 1 (2022), arXiv:2007.00664 [hep-ph]

    work page arXiv 2022

  55. [55]

    Blinov, E

    N. Blinov, E. Kowalczyk, and M. Wynne, JHEP02, 036 (2022), arXiv:2112.09814 [hep-ph]

    work page arXiv 2022

  56. [56]

    Cincheza et al., Nucl

    J. Cincheza et al., Nucl. Phys. B158, 280 (1979)

    work page 1979

  57. [57]

    J/psi and psi' production and their normal nuclear absorption in proton-nucleus collisions at 400 GeV

    B. Alessandro et al. (NA50), Eur. Phys. J. C48, 329 (2006), arXiv:nucl-ex/0612012

    work page internal anchor Pith review Pith/arXiv arXiv 2006

  58. [58]

    J/psi production in proton-nucleus collisions at 158 and 400 GeV

    R. Arnaldi et al. (NA60), Phys. Lett. B706, 263 (2012), arXiv:1004.5523 [nucl-ex]

    work page internal anchor Pith review Pith/arXiv arXiv 2012

  59. [59]

    Kobyzev, S

    I. Kobyzev, S. J. Prince, and M. A. Brubaker, IEEE transactions on pattern analysis and machine intelligence43, 3964 (2020)

    work page 2020

  60. [60]

    Papamakarios, E

    G. Papamakarios, E. Nalisnick, D. J. Rezende, S. Mohamed, and B. Lakshminarayanan, Journal of Machine Learning Research22, 1 (2021)

    work page 2021

  61. [61]

    E. G. Tabak and E. Vanden-Eijnden, Communications in Mathematical Sciences8, 217 (2010)

    work page 2010

  62. [62]

    E. G. Tabak and C. V. Turner, Communications on Pure and Applied Mathematics66, 145 (2013), https://onlinelibrary.wiley.com/doi/pdf/10.1002/cpa.21423

    work page doi:10.1002/cpa.21423 2013

  63. [63]

    L. Dinh, D. Krueger, and Y. Bengio, arXiv preprint arXiv:1410.8516 (2014)

    work page internal anchor Pith review Pith/arXiv arXiv 2014

  64. [64]

    Rezende and S

    D. Rezende and S. Mohamed, in International conference on machine learning (PMLR, 2015) pp. 1530–1538

    work page 2015

  65. [65]

    Winkler, D

    C. Winkler, D. Worrall, E. Hoogeboom, and M. Welling, (2019), arXiv:1912.00042 [cs.LG]

    work page arXiv 2019

  66. [66]

    A Living Review of Machine Learning for Particle Physics,

    HEP ML Community, “A Living Review of Machine Learning for Particle Physics,”

    work page

  67. [67]

    Stienen and R

    B. Stienen and R. Verheyen, SciPost Phys.10, 038 (2021), arXiv:2011.13445 [hep-ph]

    work page arXiv 2021

  68. [68]

    Bothmann, T

    E. Bothmann, T. Janßen, M. Knobbe, T. Schmale, and S. Schumann, SciPost Phys.8, 069 (2020), arXiv:2001.05478 [hep-ph]

    work page arXiv 2020

  69. [69]

    C. Gao, J. Isaacson, and C. Krause, Mach. Learn. Sci. Tech.1, 045023 (2020), arXiv:2001.05486 [physics.comp-ph]

    work page arXiv 2020

  70. [70]

    C. Gao, S. H¨ oche, J. Isaacson, C. Krause, and H. Schulz, Phys. Rev. D101, 076002 (2020), arXiv:2001.10028 [hep-ph]

    work page arXiv 2020

  71. [71]

    Heimel, R

    T. Heimel, R. Winterhalder, A. Butter, J. Isaacson, C. Krause, F. Maltoni, O. Mattelaer, and T. Plehn, SciPost Phys. 15, 141 (2023), arXiv:2212.06172 [hep-ph]

    work page arXiv 2023

  72. [72]

    Deutschmann and N

    N. Deutschmann and N. G¨ otz, JHEP03, 083 (2024), arXiv:2401.09069 [hep-ph]

    work page arXiv 2024

  73. [73]

    Bothmann, T

    E. Bothmann, T. Janßen, M. Knobbe, B. Schmitzer, and F. Sinz, (2025), arXiv:2506.18987 [hep-ph]

    work page arXiv 2025

  74. [74]

    Krause and D

    C. Krause and D. Shih, Phys. Rev. D107, 113003 (2023), arXiv:2106.05285 [physics.ins-det]

    work page arXiv 2023

  75. [75]

    Krause and D

    C. Krause and D. Shih, Phys. Rev. D107, 113004 (2023), arXiv:2110.11377 [physics.ins-det]

    work page arXiv 2023

  76. [76]

    Stimper, D

    V. Stimper, D. Liu, A. Campbell, V. Berenz, L. Ryll, B. Sch¨ olkopf, and J. M. Hern´ andez-Lobato, Journal of Open Source Software8, 5361 (2023)

    work page 2023

  77. [77]

    Ansel, E

    J. Ansel, E. Yang, H. He, N. Gimelshein, A. Jain, M. Voznesensky, B. Bao, P. Bell, D. Berard, E. Burovski, G. Chauhan, A. Chourdia, W. Constable, A. Desmaison, Z. DeVito, E. Ellison, W. Feng, J. Gong, M. Gschwind, B. Hirsh, S. Huang, K. Kalambarkar, L. Kirsch, M. Lazos, M. Lezcano, Y. Liang, J. Liang, Y. Lu, C. K. Luk, B. Maher, Y. Pan, C. Puhrsch, M. Res...

    work page 2024

  78. [78]

    & Papamakarios, G

    C. Durkan, A. Bekasov, I. Murray, and G. Papamakarios, inAdvances in Neural Information Processing Systems 32 (NeurIPS 2019) 13 (2019) 1906.04032

    work page arXiv 2019

  79. [79]

    Germain, K

    M. Germain, K. Gregor, I. Murray, and H. Larochelle, in International conference on machine learning (PMLR, 2015) pp. 881–889

    work page 2015

  80. [80]

    K. P. Burnham and D. R. Anderson,Model selection and multimodel inference: a practical information-theoretic approach (Springer, 2002)

    work page 2002

Showing first 80 references.