pith. sign in

arxiv: 2607.01354 · v1 · pith:COOZPA3Dnew · submitted 2026-07-01 · ✦ hep-ph

Local Conformal Predictions for Calibrated Surrogates

Suprio Dubey , Henning Bahl , Anja Butter , J\"urgen Hesser , Tilman Plehn This is my paper

Pith reviewed 2026-07-03 19:27 UTC · model grok-4.3

classification ✦ hep-ph
keywords conformal predictionneural network surrogatesLHC amplitudesuncertainty quantificationlocal calibrationscattering amplitudesevent generation
0
0 comments X

The pith

FALCON learns locally calibrated confidence intervals for neural network surrogates of LHC amplitudes via conformal prediction.

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

Neural network surrogates for LHC scattering amplitudes need reliable uncertainty estimates despite non-Gaussian errors. Standard conformal prediction methods fail to deliver locally calibrated uncertainties. The paper introduces FALCON as a new method that achieves local calibration in a distribution-free way. This matters because trustworthy fast simulations are essential for LHC physics analyses. If correct, it enables ultra-fast event generation with proper error bars.

Core claim

The authors show that standard conformal predictions struggle to provide locally calibrated uncertainties for these surrogates and introduce FALCON, a novel conformal prediction method that learns locally calibrated confidence intervals to address this.

What carries the argument

FALCON, a conformal prediction method that learns locally calibrated confidence intervals as a post-processing step on trained neural network surrogates.

Load-bearing premise

That a distribution-free post-processing step can achieve local calibration for the specific non-Gaussian systematics present in LHC scattering amplitude surrogates.

What would settle it

A test on independent LHC amplitude data where FALCON's confidence intervals show poor local coverage would disprove the method's effectiveness.

Figures

Figures reproduced from arXiv: 2607.01354 by Anja Butter, Henning Bahl, J\"urgen Hesser, Suprio Dubey, Tilman Plehn.

Figure 1
Figure 1. Figure 1: Invariant mass and scattering angle distribution for the [PITH_FULL_IMAGE:figures/full_fig_p008_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: Accuracies of qq¯ → Z g amplitude surrogates trained with a heteroscedastic loss and with a pinball loss. approaches, both roughly following a Gaussian centered around 10−5 . Towards smaller de￾viations, the tails look identical for both approaches, whereas towards larger deviations the quantile regression processing leads to a slightly more suppressed tail. Marginal coverage To test the coverage, we add G… view at source ↗
Figure 3
Figure 3. Figure 3: Marginal versus nominal coverage for decreasing levels of artificial Gaus [PITH_FULL_IMAGE:figures/full_fig_p011_3.png] view at source ↗
Figure 4
Figure 4. Figure 4: Conditional 68% CL coverage as a function of [PITH_FULL_IMAGE:figures/full_fig_p012_4.png] view at source ↗
Figure 5
Figure 5. Figure 5: Conditional coverage from FALCON-CQR for different CLs as a function of [PITH_FULL_IMAGE:figures/full_fig_p012_5.png] view at source ↗
Figure 6
Figure 6. Figure 6: Phase-space 68% CL coverage in the (mZ g , cosθ) plane, without artificial noise. The black lines in the FALCON panel indicate the probe windows at fixed mZ g . 13 [PITH_FULL_IMAGE:figures/full_fig_p013_6.png] view at source ↗
Figure 7
Figure 7. Figure 7: Left: invariant mass distribution mZ g g of qq¯ → Z g g. Right: relative accu￾racy of the qq¯ → Z g g amplitude surrogates trained with a heteroscedastic loss and with a pinball loss. 14 [PITH_FULL_IMAGE:figures/full_fig_p014_7.png] view at source ↗
Figure 8
Figure 8. Figure 8: Left: marginal versus nominal coverage for [PITH_FULL_IMAGE:figures/full_fig_p015_8.png] view at source ↗
Figure 9
Figure 9. Figure 9: Conditional coverage from FALCON for different CLs as a function of [PITH_FULL_IMAGE:figures/full_fig_p016_9.png] view at source ↗
Figure 10
Figure 10. Figure 10: Marginal coverage vs. nominal coverage for FALCON (CQR) for [PITH_FULL_IMAGE:figures/full_fig_p018_10.png] view at source ↗
Figure 11
Figure 11. Figure 11: Coverage of FALCON (CQR) for qq¯ → Z g with varying nfresh with ncalib = 5000 fixed. Left: marginal coverage vs. nominal coverage, with the dashed diagonal representing exact coverage. Right: Conditional 68% CL coverage as a func￾tion of mZ g . The marginal guarantee holds for as few as nfresh = 10 fresh probe events. Conditional coverage fluctuates at nfresh = 10 and stabilises near nominal for nfresh ≳ … view at source ↗
Figure 12
Figure 12. Figure 12: Conditional 68% CL coverage for qq¯ → Z g as a function of cosθ. FALCON is calibrated at 10 evenly spaced probe windows along mZ g . For the qq¯ → Z g case, we choose probe regions as bins in mZ g ( [PITH_FULL_IMAGE:figures/full_fig_p019_12.png] view at source ↗
Figure 13
Figure 13. Figure 13: FALCON with 2D probe interpolation for qq¯ → Z g at the 1σ level (1 − α ≈ 0.68). The phase-space coverage across the (mZ g , cosθ) plane is shown. The 11 probe regions are indicated by the black boxes. As an alternative to selecting mZ g bins as probe regions for FALCON-CQR — as shown in [PITH_FULL_IMAGE:figures/full_fig_p020_13.png] view at source ↗
Figure 14
Figure 14. Figure 14: Conditional 68% CL coverage from FALCON-CQR with 2D probe interpo [PITH_FULL_IMAGE:figures/full_fig_p020_14.png] view at source ↗
Figure 15
Figure 15. Figure 15: Conditional 68% CL coverage for qq¯ → Z g as a function of mZ g , for ϵ = 10−3 (left), ϵ = 10−4 (center), and ϵ = 10−5 (right) applied at mthr = 200 GeV according to Eq. (42). 0.0 0.2 0.4 0.6 0.8 1.0 Nominal coverage (1 − α) 0.0 0.2 0.4 0.6 0.8 1.0 Marginal coverage " = 10−3 mthreshold = 200 GeV Gaussian QR CP Het. CQR FALCON-CQR 0.0 0.2 0.4 0.6 0.8 1.0 Nominal coverage (1 − α) 0.0 0.2 0.4 0.6 0.8 1.0 Mar… view at source ↗
Figure 16
Figure 16. Figure 16: Marginal coverage vs nominal coverage for [PITH_FULL_IMAGE:figures/full_fig_p021_16.png] view at source ↗
Figure 17
Figure 17. Figure 17: Pull distributions for two mZ g slices based on the qq¯ → Z g heteroscedastic surrogate. To further illustrate the Gaussian coverage observations of Figs. 4 and 6, we investigate the pull t(x) = ANN(x) − Atrue(x) σ(x) . (43) If the residuals are Gaussian distributed and σ is correctly calibrated, the pull distribution will follow a unit Gaussian. We investigate it in two mZ g slices. For mZ g ≈ 113 ... 15… view at source ↗
read the original abstract

Neural network surrogates for LHC scattering amplitudes require trustworthy uncertainty estimates, a challenging task given the non-Gaussian systematics. We target it using conformal prediction, a distribution-free post-processing to complement trained surrogates with calibrated uncertainties. We find that standard conformal predictions struggle to provide locally calibrated uncertainties. This leads us to introduce FALCON, a novel conformal prediction method that learns locally calibrated confidence intervals. Our simple examples illustrate the power of distribution-free uncertainty quantification for ultra-fast event generation at the LHC.

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 paper claims that standard conformal prediction methods fail to deliver locally calibrated uncertainties for neural network surrogates of LHC scattering amplitudes due to non-Gaussian systematics. It introduces FALCON as a novel distribution-free post-processing method that learns locally calibrated confidence intervals, with the approach illustrated on simple examples for applications to ultra-fast event generation at the LHC.

Significance. If the central construction holds, the work offers a potentially useful addition to uncertainty quantification tools in high-energy physics ML surrogates, where local calibration matters for reliable amplitude predictions. The explicit distribution-free framing and focus on local rather than marginal calibration are strengths that align with practical needs in LHC phenomenology.

major comments (2)
  1. [§3] §3 (method definition): the claim that FALCON achieves local calibration via a distribution-free step requires explicit verification that the learned intervals remain valid under the non-Gaussian error structures typical of LHC amplitude surrogates; the simple examples do not test this regime and therefore do not establish the result for the stated target application.
  2. [§4] §4 (experiments): the reported calibration metrics on toy examples show improvement over standard conformal prediction, but no quantitative comparison is given for coverage as a function of local density or for the specific non-Gaussian tails mentioned in the abstract; without these controls the local-calibration advantage remains unproven for the LHC use case.
minor comments (2)
  1. Notation for the local conformity score is introduced without a clear comparison table to the standard conformal score; adding such a table would improve readability.
  2. The abstract states the method is 'distribution-free' yet the training of the local adjustment appears to require a held-out calibration set whose size is not discussed; a brief remark on sample complexity would help.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for the constructive comments. We respond point-by-point to the major comments below.

read point-by-point responses

  1. Referee: [§3] §3 (method definition): the claim that FALCON achieves local calibration via a distribution-free step requires explicit verification that the learned intervals remain valid under the non-Gaussian error structures typical of LHC amplitude surrogates; the simple examples do not test this regime and therefore do not establish the result for the stated target application.

    Authors: FALCON is constructed as a distribution-free procedure that inherits coverage guarantees from conformal prediction theory; these guarantees hold for arbitrary error distributions and do not require Gaussianity. The local calibration step preserves the necessary exchangeability conditions. The simple examples are intended to illustrate the local behavior in controlled settings. We will revise §3 to add an explicit statement of the validity conditions together with a short argument showing that the guarantees are independent of the error distribution. revision: partial

  2. Referee: [§4] §4 (experiments): the reported calibration metrics on toy examples show improvement over standard conformal prediction, but no quantitative comparison is given for coverage as a function of local density or for the specific non-Gaussian tails mentioned in the abstract; without these controls the local-calibration advantage remains unproven for the LHC use case.

    Authors: The current experiments report improved local calibration metrics on the toy examples. We agree that explicit quantitative controls for coverage versus local density and for non-Gaussian tails would strengthen the connection to the LHC application. In the revised manuscript we will add these comparisons, including plots of empirical coverage as a function of local density and additional runs with heavy-tailed noise models. revision: yes

Circularity Check

0 steps flagged

No significant circularity identified

full rationale

The abstract and visible text introduce FALCON as a novel post-processing method for local calibration in conformal prediction applied to LHC surrogates, but contain no equations, derivations, fitted parameters, or self-citations. The distinction between standard conformal prediction and FALCON is presented as an empirical observation without any reduction of a claimed prediction or uniqueness result to the paper's own inputs by construction. The central claim remains an independent methodological proposal whose validity would be tested externally rather than forced internally.

Axiom & Free-Parameter Ledger

0 free parameters · 0 axioms · 0 invented entities

Abstract provides no information on free parameters, axioms, or invented entities; all such elements are unknown.

pith-pipeline@v0.9.1-grok · 5611 in / 948 out tokens · 17184 ms · 2026-07-03T19:27:47.966211+00:00 · methodology

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

298 extracted references · 290 canonical work pages · 3 internal anchors

  1. [1]

    MadSpace -- Event Generation for the Era of GPUs and ML

    Heimel, Theo and Mattelaer, Olivier and Winterhalder, Ramon. MadSpace -- Event Generation for the Era of GPUs and ML. 2026. arXiv:2602.06895

    work page arXiv 2026

  2. [2]

    Monte Carlo Event Generation with Continuous Normalizing Flows

    Bothmann, Enrico and Jan en, Timo and Knobbe, Max and Schmitzer, Bernhard and Sinz, Fabian. Monte Carlo Event Generation with Continuous Normalizing Flows. 2026. arXiv:2604.03511

    work page internal anchor Pith review Pith/arXiv arXiv 2026

  3. [3]

    A Gentle Introduction to Conformal Prediction and Distribution-Free Uncertainty Quantification

    Anastasios N. Angelopoulos and Stephen Bates , title =. CoRR , volume =. 2021 , url =. 2107.07511 , timestamp =

    work page internal anchor Pith review Pith/arXiv arXiv 2021

  4. [4]

    Conformalized Quantile Regression , volume =

    Romano, Yaniv and Patterson, Evan and Candes, Emmanuel , booktitle =. Conformalized Quantile Regression , volume =. 2019 , eprint =

    2019

  5. [5]

    2024 , MONTH = Sep, PDF =

    Da Veiga, S. 2024 , MONTH = Sep, PDF =

    2024

  6. [6]

    Tibshirani , title =

    Ryan J. Tibshirani , title =. 2023 , howpublished =

    2023

  7. [7]

    Madgraph5\_aMC@NLO on GPUs and vector CPUs Experience with the first alpha release

    Wettersten, Zenny and Mattelaer, Olivier and Roiser, Stefan and Sch\"ofbeck, Robert and Valassi, Andrea , title = ". EPJ Web Conf. 2024. doi:10.1051/epjconf/202429510001. arXiv:2312.07440

    work page doi:10.1051/epjconf/202429510001 2024

  8. [8]

    Madgraph5\_aMC@NLO on GPUs and vector CPUs Experience with the first alpha release

    Hageboeck, Stephan and Childers, Taylor and Hopkins, Walter and Mattelaer, Olivier and Nichols, Nathan and Roiser, Stefan and Teig, J rgen and Valassi, Andrea and Vuosalo, Carl and Wettersten, Zenny. Madgraph5\_aMC@NLO on GPUs and vector CPUs Experience with the first alpha release. EPJ Web Conf. 2024. doi:10.1051/epjconf/202429511013. arXiv:2312.02898

    work page doi:10.1051/epjconf/202429511013 2024

  9. [9]

    and others

    Valassi, A. and others. Speeding up Madgraph5 aMC@NLO through CPU vectorization and GPU offloading: towards a first alpha release. 21th International Workshop on Advanced Computing and Analysis Techniques in Physics Research : AI meets Reality. 2023. arXiv:2303.18244

    work page arXiv 2023

  10. [10]

    oche, Stefan and Isaacson, Joshua and Knobbe, Max , title =

    Bothmann, Enrico and Childers, Taylor and Giele, Walter and H\"oche, Stefan and Isaacson, Joshua and Knobbe, Max , title = ". SciPost Phys. 2024. doi:10.21468/SciPostPhys.17.3.081. arXiv:2311.06198

    work page doi:10.21468/scipostphys.17.3.081 2024

  11. [11]

    utschow, Christian and H\

    Bothmann, Enrico and Buckley, Andy and Christidi, Ilektra A. and G\"utschow, Christian and H\"oche, Stefan and Knobbe, Max and Martin, Tim and Sch\"onherr, Marek , title = ". Eur. Phys. J. C. 2022. doi:10.1140/epjc/s10052-022-11087-1. arXiv:2209.00843

    work page doi:10.1140/epjc/s10052-022-11087-1 2022

  12. [12]

    Branches of a Tree: Taking Derivatives of Programs with Discrete and Branching Randomness in High Energy Physics

    Kagan, Michael and Heinrich, Lukas. Branches of a Tree: Taking Derivatives of Programs with Discrete and Branching Randomness in High Energy Physics. 2023. arXiv:2308.16680

    work page arXiv 2023

  13. [13]

    MadFlow: automating Monte Carlo simulation on GPU for particle physics processes

    Carrazza, Stefano and Cruz-Martinez, Juan and Rossi, Marco and Zaro, Marco. MadFlow: automating Monte Carlo simulation on GPU for particle physics processes. Eur. Phys. J. C. 2021. doi:10.1140/epjc/s10052-021-09443-8. arXiv:2106.10279

    work page doi:10.1140/epjc/s10052-021-09443-8 2021

  14. [14]

    and Rossi, Marco

    Carrazza, Stefano and Cruz-Martinez, Juan M. and Rossi, Marco. PDFFlow: Parton distribution functions on GPU. Comput. Phys. Commun. 2021. doi:10.1016/j.cpc.2021.107995. arXiv:2009.06635

    work page doi:10.1016/j.cpc.2021.107995 2021

  15. [15]

    otz, Niklas , title =

    Deutschmann, Nicolas and G\"otz, Niklas , title = ". JHEP. 2024. doi:10.1007/JHEP03(2024)083. arXiv:2401.09069

    work page doi:10.1007/jhep03(2024)083 2024

  16. [16]

    Unifying Simulation and Inference with Normalizing Flows

    Du, Haoxing and Krause, Claudius and Mikuni, Vinicius and Nachman, Benjamin and Pang, Ian and Shih, David. Unifying Simulation and Inference with Normalizing Flows. 2024. arXiv:2404.18992

    work page arXiv 2024

  17. [17]

    Convolutional L2LFlows: Generating Accurate Showers in Highly Granular Calorimeters Using Convolutional Normalizing Flows

    Buss, Thorsten and Gaede, Frank and Kasieczka, Gregor and Krause, Claudius and Shih, David. Convolutional L2LFlows: Generating Accurate Showers in Highly Granular Calorimeters Using Convolutional Normalizing Flows. 2024. arXiv:2405.20407

    work page arXiv 2024

  18. [18]

    PIPPIN: Generating variable length full events from partons

    Qu\'etant, Guillaume and Raine, John Andrew and Leigh, Matthew and Sengupta, Debajyoti and Golling, Tobias. PIPPIN: Generating variable length full events from partons. 2024. arXiv:2406.13074

    work page arXiv 2024

  19. [19]

    Toward the end-to-end optimization of particle physics instruments with differentiable programming

    Dorigo, Tommaso and others. Toward the end-to-end optimization of particle physics instruments with differentiable programming. Rev. Phys. 2023. doi:10.1016/j.revip.2023.100085. arXiv:2203.13818

    work page doi:10.1016/j.revip.2023.100085 2023

  20. [20]

    Morphing parton showers with event derivatives

    Nachman, Benjamin and Prestel, Stefan. Morphing parton showers with event derivatives. 2022. arXiv:2208.02274

    work page arXiv 2022

  21. [21]

    and Heinrich, Lukas and Kagan, Michael and Lange, David

    Aehle, Max and Nov\'ak, Mih\'aly and Vassilev, Vassil and Gauger, Nicolas R. and Heinrich, Lukas and Kagan, Michael and Lange, David. Optimization Using Pathwise Algorithmic Derivatives of Electromagnetic Shower Simulations. 2024. arXiv:2405.07944

    work page arXiv 2024

  22. [22]

    Whalley, M. R. and Bourilkov, D. and Group, R. C. The Les Houches accord PDFs (LHAPDF) and LHAGLUE. HERA and the LHC: A Workshop on the Implications of HERA and LHC Physics (Startup Meeting, CERN, 26-27 March 2004; Midterm Meeting, CERN, 11-13 October 2004). 2005. arXiv:hep-ph/0508110

    work page internal anchor Pith review Pith/arXiv arXiv 2004

  23. [23]

    Unweighting multijet event generation using factorisation-aware neural networks

    Jan en, Timo and Ma\^ tre, Daniel and Schumann, Steffen and Siegert, Frank and Truong, Henry. Unweighting multijet event generation using factorisation-aware neural networks. SciPost Phys. 2023. doi:10.21468/SciPostPhys.15.3.107. arXiv:2301.13562

    work page doi:10.21468/scipostphys.15.3.107 2023

  24. [24]

    Exploring phase space with Nested Sampling

    Yallup, David and Jan en, Timo and Schumann, Steffen and Handley, Will. Exploring phase space with Nested Sampling. Eur. Phys. J. C. 2022. doi:10.1140/epjc/s10052-022-10632-2. arXiv:2205.02030

    work page doi:10.1140/epjc/s10052-022-10632-2 2022

  25. [25]

    Efficient phase-space generation for hadron collider event simulation

    Bothmann, Enrico and Childers, Taylor and Giele, Walter and Herren, Florian and Hoeche, Stefan and Isaacsson, Joshua and Knobbe, Max and Wang, Rui. Efficient phase-space generation for hadron collider event simulation. SciPost Phys. 2023. doi:10.21468/SciPostPhys.15.4.169. arXiv:2302.10449

    work page doi:10.21468/scipostphys.15.4.169 2023

  26. [26]

    Event-by-event Comparison between Machine-Learning- and Transfer-Matrix-based Unfolding Methods

    Backes, Mathias and Butter, Anja and Dunford, Monica and Malaescu, Bogdan. Event-by-event Comparison between Machine-Learning- and Transfer-Matrix-based Unfolding Methods. 2023. arXiv:2310.17037

    work page arXiv 2023

  27. [27]

    Faster diffusion model with improved quality for particle cloud generation

    Leigh, Matthew and Sengupta, Debajyoti and Raine, John Andrew and Qu\'etant, Guillaume and Golling, Tobias. Faster diffusion model with improved quality for particle cloud generation. Phys. Rev. D. 2024. doi:10.1103/PhysRevD.109.012010. arXiv:2307.06836

    work page doi:10.1103/physrevd.109.012010 2024

  28. [28]

    Fast and improved neutrino reconstruction in multineutrino final states with conditional normalizing flows

    Raine, John Andrew and Leigh, Matthew and Zoch, Knut and Golling, Tobias. Fast and improved neutrino reconstruction in multineutrino final states with conditional normalizing flows. Phys. Rev. D. 2024. doi:10.1103/PhysRevD.109.012005. arXiv:2307.02405

    work page doi:10.1103/physrevd.109.012005 2024

  29. [29]

    and Golling, Tobias and Kasieczka, Gregor and Leigh, Matthew and Qu\'etant, Guillaume and Raine, John Andrew and Sengupta, Debajyoti and Shih, David

    Buhmann, Erik and Ewen, Cedric and Faroughy, Darius A. and Golling, Tobias and Kasieczka, Gregor and Leigh, Matthew and Qu\'etant, Guillaume and Raine, John Andrew and Sengupta, Debajyoti and Shih, David. EPiC-ly Fast Particle Cloud Generation with Flow-Matching and Diffusion. 2023. arXiv:2310.00049

    work page arXiv 2023

  30. [30]

    2308.00027 , archiveprefix =

    Ackerschott, Jona and Barman, Rahool Kumar and Gon. 2308.00027 , archiveprefix =

    work page arXiv

  31. [31]

    SciPost Phys

    Badger, Simon and Butter, Anja and Luchmann, Michel and Pitz, Sebastian and Plehn, Tilman , title =. SciPost Phys. Core , volume = 6, pages =. doi:10.21468/SciPostPhysCore.6.2.034 , eprint =

    work page doi:10.21468/scipostphyscore.6.2.034

  32. [32]

    SciPost Phys

    Badger, Simon and others , title =. SciPost Phys. , booktitle =. doi:10.21468/SciPostPhys.14.4.079 , editor =. 2203.07460 , archiveprefix =

    work page doi:10.21468/scipostphys.14.4.079

  33. [33]

    and Krause, Claudius and Pang, Ian and Shih, David

    Buckley, Matthew R. and Krause, Claudius and Pang, Ian and Shih, David. Inductive simulation of calorimeter showers with normalizing flows. Phys. Rev. D. 2024. doi:10.1103/PhysRevD.109.033006. arXiv:2305.11934

    work page doi:10.1103/physrevd.109.033006 2024

  34. [34]

    uger, Katja and McKeown, Peter , title =

    Buhmann, Erik and Diefenbacher, Sascha and Eren, Engin and Gaede, Frank and Kasieczka, Gregor and Korol, Anatolii and Korcari, William and Kr\"uger, Katja and McKeown, Peter , title = ". JINST. 2023. doi:10.1088/1748-0221/18/11/P11025. arXiv:2305.04847

    work page doi:10.1088/1748-0221/18/11/p11025 2023

  35. [35]

    EPiC-GAN: Equivariant point cloud generation for particle jets

    Buhmann, Erik and Kasieczka, Gregor and Thaler, Jesse. EPiC-GAN: Equivariant point cloud generation for particle jets. SciPost Phys. 2023. doi:10.21468/SciPostPhys.15.4.130. arXiv:2301.08128

    work page doi:10.21468/scipostphys.15.4.130 2023

  36. [36]

    SciPost Phys

    Butter, Anja and Heimel, Theo and Hummerich, Sander and Krebs, Tobias and Plehn, Tilman and Rousselot, Armand and Vent, Sophia , title =. SciPost Phys. , volume = 14, number = 4, pages =. doi:10.21468/SciPostPhys.14.4.078 , eprint =

    work page doi:10.21468/scipostphys.14.4.078

  37. [37]

    2305.10475 , archiveprefix =

    Butter, Anja and Huetsch, Nathan and Schweitzer, Sofia Palacios and Plehn, Tilman and Sorrenson, Peter and Spinner, Jonas , title =. 2305.10475 , archiveprefix =

    work page arXiv

  38. [38]

    2302.12024 , archiveprefix =

    Coccaro, Andrea and Letizia, Marco and Reyes-Gonzalez, Humberto and Torre, Riccardo , title =. 2302.12024 , archiveprefix =

    work page arXiv

  39. [39]

    2305.16774 , archiveprefix =

    Das, Ranit and Favaro, Luigi and Heimel, Theo and Krause, Claudius and Plehn, Tilman and Shih, David , title =. 2305.16774 , archiveprefix =

    work page arXiv

  40. [40]

    uger, Katja and McKeown, Peter and Rustige, Lennart , title =

    Diefenbacher, Sascha and Eren, Engin and Gaede, Frank and Kasieczka, Gregor and Korol, Anatolii and Kr\"uger, Katja and McKeown, Peter and Rustige, Lennart , title = ". Mach. Learn. Sci. Tech. 2023. doi:10.1088/2632-2153/acefa9. arXiv:2303.18150

    work page doi:10.1088/2632-2153/acefa9 2023

  41. [41]

    L2LFlows: generating high-fidelity 3D calorimeter images

    Diefenbacher, Sascha and Eren, Engin and Gaede, Frank and Kasieczka, Gregor and Krause, Claudius and Shekhzadeh, Imahn and Shih, David. L2LFlows: generating high-fidelity 3D calorimeter images. JINST. 2023. doi:10.1088/1748-0221/18/10/P10017. arXiv:2302.11594

    work page doi:10.1088/1748-0221/18/10/p10017 2023

  42. [42]

    2308.12351 , archiveprefix =

    Diefenbacher, Sascha and Liu, Guan-Horng and Mikuni, Vinicius and Nachman, Benjamin and Nie, Weili , title =. 2308.12351 , archiveprefix =

    work page arXiv

  43. [43]

    2308.12339 , archiveprefix =

    Diefenbacher, Sascha and Mikuni, Vinicius and Nachman, Benjamin , title =. 2308.12339 , archiveprefix =

    work page arXiv

  44. [44]

    and Favaro, Luigi and Feiden, Friedrich and Modak, Tanmoy and Plehn, Tilman , title =

    Dillon, Barry M. and Favaro, Luigi and Feiden, Friedrich and Modak, Tanmoy and Plehn, Tilman , title =. 2301.04660 , archiveprefix =

    work page arXiv

  45. [45]

    2308.16716 , archiveprefix =

    Dittmaier, Stefan and Maierh\"ofer, Philipp and Schwan, Christopher and Winterhalder, Ramon , title =. 2308.16716 , archiveprefix =

    work page arXiv

  46. [46]

    23xx.xxxx , archiveprefix =

    Ernst, Florian and Favaro, Luigi and Krause, Claudius and Plehn, Tilman and Shih, David , title =. 23xx.xxxx , archiveprefix =

  47. [47]

    Ultra-high-granularity detector simulation with intra-event aware generative adversarial network and self-supervised relational reasoning

    Hashemi, Baran and Hartmann, Nikolai and Sharifzadeh, Sahand and Kahn, James and Kuhr, Thomas. Ultra-high-granularity detector simulation with intra-event aware generative adversarial network and self-supervised relational reasoning. Nature Commun. 2024. doi:10.1038/s41467-024-49104-4. arXiv:2303.08046

    work page doi:10.1038/s41467-024-49104-4 2024

  48. [48]

    Kansal, Raghav and Li, Anni and Duarte, Javier and Chernyavskaya, Nadezda and Pierini, Maurizio and Orzari, Breno and Tomei, Thiago , title =. Phys. Rev. D , volume = 107, number = 7, pages =. doi:10.1103/PhysRevD.107.076017 , eprint =

    work page doi:10.1103/physrevd.107.076017

  49. [49]

    Kasieczka, Gregor and Mastandrea, Radha and Mikuni, Vinicius and Nachman, Benjamin and Pettee, Mariel and Shih, David , title =. Phys. Rev. D , volume = 107, number = 1, pages =. doi:10.1103/PhysRevD.107.015009 , eprint =

    work page doi:10.1103/physrevd.107.015009

  50. [50]

    PC-JeDi: Diffusion for particle cloud generation in high energy physics

    Leigh, Matthew and Sengupta, Debajyoti and Qu\'etant, Guillaume and Raine, John Andrew and Zoch, Knut and Golling, Tobias. PC-JeDi: Diffusion for particle cloud generation in high energy physics. SciPost Phys. 2024. doi:10.21468/SciPostPhys.16.1.018. arXiv:2303.05376

    work page doi:10.21468/scipostphys.16.1.018 2024

  51. [51]

    M\"uhlleitner, Margarete and Plehn, Tilman , title =

  52. [52]

    and Truong, H

    Ma\^ tre, D. and Truong, H. One-loop matrix element emulation with factorisation awareness. JHEP , volume =. 2023. doi:10.1007/JHEP05(2023)159. arXiv:2302.04005

    work page doi:10.1007/jhep05(2023)159 2023

  53. [53]

    Fast point cloud generation with diffusion models in high energy physics

    Mikuni, Vinicius and Nachman, Benjamin and Pettee, Mariel. Fast point cloud generation with diffusion models in high energy physics. Phys. Rev. D. 2023. doi:10.1103/PhysRevD.108.036025. arXiv:2304.01266

    work page doi:10.1103/physrevd.108.036025 2023

  54. [54]

    Elsa: enhanced latent spaces for improved collider simulations

    Nachman, Benjamin and Winterhalder, Ramon. Elsa: enhanced latent spaces for improved collider simulations. Eur. Phys. J. C. 2023. doi:10.1140/epjc/s10052-023-11989-8. arXiv:2305.07696

    work page doi:10.1140/epjc/s10052-023-11989-8 2023

  55. [55]

    Raine, John Andrew and Klein, Samuel and Sengupta, Debajyoti and Golling, Tobias , title =. Front. Big Data , volume = 6, pages = 899345, doi =. 2203.09470 , archiveprefix =

    work page arXiv

  56. [56]

    2305.10500 , archiveprefix =

    Rizvi, Shahzar and Pettee, Mariel and Nachman, Benjamin , title =. 2305.10500 , archiveprefix =

    work page arXiv

  57. [57]

    2305.04646 , archiveprefix =

    Sengupta, Debajyoti and Klein, Samuel and Raine, John Andrew and Golling, Tobias , title =. 2305.04646 , archiveprefix =

    work page arXiv

  58. [58]

    2305.10399 , archiveprefix =

    Shmakov, Alexander and Greif, Kevin and Fenton, Michael and Ghosh, Aishik and Baldi, Pierre and Whiteson, Daniel , title =. 2305.10399 , archiveprefix =

    work page arXiv

  59. [59]

    Generative machine learning for detector response modeling with a conditional normalizing flow

    Xu, Allison and Han, Shuo and Ju, Xiangyang and Wang, Haichen. Generative machine learning for detector response modeling with a conditional normalizing flow. JINST. 2024. doi:10.1088/1748-0221/19/02/P02003. arXiv:2303.10148

    work page doi:10.1088/1748-0221/19/02/p02003 2024

  60. [60]

    Deep Generative Models for Fast Photon Shower Simulation in ATLAS. Comput. Softw. Big Sci. 2024. doi:10.1007/s41781-023-00106-9. arXiv:2210.06204

    work page doi:10.1007/s41781-023-00106-9 2024

  61. [61]

    Comput. Softw. Big Sci. , volume = 6, pages = 7, doi =. 2109.02551 , archiveprefix =

    work page arXiv

  62. [62]

    2203.08806 , archiveprefix =

    Adelmann, Andreas and others , title =. 2203.08806 , archiveprefix =

    work page arXiv

  63. [63]

    Albouy, Guillaume and others , title =. Eur. Phys. J. C , volume = 82, number = 12, pages = 1132, doi =. 2203.09503 , archiveprefix =

    work page arXiv

  64. [64]

    JINST , volume = 17, number =

    Arratia, Miguel and others , title =. JINST , volume = 17, number =. doi:10.1088/1748-0221/17/01/P01024 , eprint =

    work page doi:10.1088/1748-0221/17/01/p01024

  65. [65]

    2212.08674 , archiveprefix =

    Backes, Mathias and Butter, Anja and Dunford, Monica and Malaescu, Bogdan , title =. 2212.08674 , archiveprefix =

    work page arXiv

  66. [66]

    and Keilbach, Fabian and Plehn, Tilman and Kasieczka, Gregor and Whiteson, Daniel , title =

    Baldi, Pierre and Blecher, Lukas and Butter, Anja and Collado, Julian and Howard, Jessica N. and Keilbach, Fabian and Plehn, Tilman and Kasieczka, Gregor and Whiteson, Daniel , title =. SciPost Phys. , volume = 13, number = 3, pages =. doi:10.21468/SciPostPhys.13.3.064 , eprint =

    work page doi:10.21468/scipostphys.13.3.064

  67. [68]

    SciPost Phys

    Bellagente, Marco and Haussmann, Manuel and Luchmann, Michel and Plehn, Tilman , title =. SciPost Phys. , volume = 13, number = 1, pages =. doi:10.21468/SciPostPhys.13.1.003 , eprint =

    work page doi:10.21468/scipostphys.13.1.003

  68. [69]

    Benato, Lisa and others , title =. Comput. Softw. Big Sci. , volume = 6, number = 1, pages = 9, doi =. 2107.00656 , archiveprefix =

    work page arXiv

  69. [70]

    JINST , volume = 17, number =

    Bieringer, Sebastian and Butter, Anja and Diefenbacher, Sascha and Eren, Engin and Gaede, Frank and Hundhausen, Daniel and Kasieczka, Gregor and Nachman, Benjamin and Plehn, Tilman and Trabs, Mathias , title =. JINST , volume = 17, number =. doi:10.1088/1748-0221/17/09/P09028 , eprint =

    work page doi:10.1088/1748-0221/17/09/p09028

  70. [71]

    2208.08454 , archiveprefix =

    Brivio, Ilaria and Bruggisser, Sebastian and Elmer, Nina and Geoffray, Emma and Luchmann, Michel and Plehn, Tilman , title =. 2208.08454 , archiveprefix =

    work page arXiv

  71. [72]

    SciPost Phys

    Brivio, Ilaria and Bruggisser, Sebastian and Geoffray, Emma and Killian, Wolfgang and Kr\"amer, Michael and Luchmann, Michel and Plehn, Tilman and Summ, Benjamin , title =. SciPost Phys. , volume = 12, number = 1, pages =. doi:10.21468/SciPostPhys.12.1.036 , eprint =

    work page doi:10.21468/scipostphys.12.1.036

  72. [73]

    Buhmann, Erik and Diefenbacher, Sascha and Hundhausen, Daniel and Kasieczka, Gregor and Korcari, William and Eren, Engin and Gaede, Frank and Kr\"uger, Katja and McKeown, Peter and Rustige, Lennart , title =. Mach. Learn. Sci. Tech. , volume = 3, number = 2, pages =. doi:10.1088/2632-2153/ac7848 , eprint =

    work page doi:10.1088/2632-2153/ac7848

  73. [74]

    amer, Michael and Morandini, Alessandro and M\

    Buss, Thorsten and Dillon, Barry M. and Finke, Thorben and Kr\"amer, Michael and Morandini, Alessandro and M\"uck, Alexander and Oleksiyuk, Ivan and Plehn, Tilman , title =. 2202.00686 , archiveprefix =

    work page arXiv

  74. [75]

    SciPost Phys

    Butter, Anja and Diefenbacher, Sascha and Kasieczka, Gregor and Nachman, Benjamin and Plehn, Tilman and Shih, David and Winterhalder, Ramon , title =. SciPost Phys. , volume = 13, number = 4, pages =. doi:10.21468/SciPostPhys.13.4.087 , eprint =

    work page doi:10.21468/scipostphys.13.4.087

  75. [76]

    and Plehn, Tilman and Vogel, Lorenz , title =

    Butter, Anja and Dillon, Barry M. and Plehn, Tilman and Vogel, Lorenz , title =. 2212.10493 , archiveprefix =

    work page arXiv

  76. [77]

    SciPost Phys

    Butter, Anja and Heimel, Theo and Martini, Till and Peitzsch, Sascha and Plehn, Tilman , title =. SciPost Phys. , volume = 15, pages =. doi:10.21468/SciPostPhys.15.3.094 , eprint =

    work page doi:10.21468/scipostphys.15.3.094

  77. [78]

    Campbell, J. M. and others , title =. 2203.11110 , archiveprefix =

    work page arXiv

  78. [79]

    , title =

    Chahrour, Ibrahim and Wells, James D. , title =. SciPost Phys. , volume = 12, pages = 187, doi =. 2111.14788 , archiveprefix =

    work page arXiv

  79. [80]

    JCAP , volume = 11, pages =

    Coogan, Adam and Morrison, Logan and Plehn, Tilman and Profumo, Stefano and Reimitz, Peter , title =. JCAP , volume = 11, pages =. doi:10.1088/1475-7516/2022/11/033 , eprint =

    work page doi:10.1088/1475-7516/2022/11/033 2022

  80. [81]

    and Ross, Brendan Leigh and Loaiza-Ganem, Gabriel and Reyes-Gonzalez, Humberto and Letizia, Marco and Caterini, Anthony L

    Cresswell, Jesse C. and Ross, Brendan Leigh and Loaiza-Ganem, Gabriel and Reyes-Gonzalez, Humberto and Letizia, Marco and Caterini, Anthony L. , title =. 2211.15380 , archiveprefix =

    work page arXiv

Showing first 80 references.