pith. sign in

arxiv: 2605.18861 · v1 · pith:7N3WP3IXnew · submitted 2026-05-14 · ⚛️ physics.ins-det · hep-ex

Enhanced Ionization Charge Identification in the Short-Baseline Neutrino Program Neutrino Detectors with Deep Neural Networks

P. Abratenko , N. Abrego-Martinez , R. Acciarri , A. Aduszkiewicz , F. Akbar , D. Andrade Aldana , L. Aliaga-Soplin , F. Abd Alrahman
show 328 more authors
R. Alvarez-Garrote C. Andreopoulos A. Antonakis M. Artero Pons J. Asaadi W. F. Badgett S. Baena B. Baibussinov S. Balasubramanian A. Barnard V. Basque J. Bateman A. Beever B. Behera E. Belchior V. Bellini R. Benocci J. Berger S. Bertolucci M. Betancourt A. Bhat M. Bishai A. Blake A. Blanchet F. Boffelli B. Bogart M. Bonesini T. Boone B. Bottino A. Braggiotti D. Brailsford A. Brandt S. J. Brice S. Brickner V. Brio C. Brizzolari M. B. Brunetti H. S. Budd L. Camilleri A. Campani A. Campos D. Caratelli D. Carber B. Carlson M. F. Carneiro I. Caro Terrazas H. Carranza R. Castillo F. Castillo Fernandez F. Cavanna S. Centro G. Cerati A. Chappell A. Chatterjee H. Chen D. Cherdack S. Cherubini N. Chithirasreemadam S. Chung M. F. Cicala M. Cicerchia R. Coackley T. E. Coan A. Cocco M. R. Convery L. Cooper-Troendle S. Copello C. Cuesta Y. Dabburi O. Dalager M. Dall'Olio A. A. Dange R. Darby S. Kr Das M. Diwan Z. Djurcic S. Dolan S. Dominguez-Vidales S. Di Domizio S. Donati F. Drielsma M. Dubnowski K. Duffy J. Dyer S. Dytman A. Ereditato J. J. Evans A. Ezeribe A. Falcone C. Fan C. Farnese A. Fava D. Di Ferdinando A. Filkins B. Fleming W. Foreman D. Franco G. Fricano I. Furic A. Furmanski N. Gallice S. Gao D. Garcia-Gamez S. Gardiner C. Gatto D. Gibin I. Gil-Botella A. Gioiosa S. Gollapinni P. Green W. C. Griffith W. Gu A. Guglielmi G. Gurung L. Hagaman P. Hamilton K. Hassinin H. Hausner A. Heggestuen A. Hergenhan M. Hernandez-Morquecho P. Holanda B. Howard R. Howell Z. Hulcher I. Ingratta M. S. Ismail C. James W. Jang R. S. Jones M. Jung T. Junk Y.-J. Jwa D. Kalra G. Karagiorgi L. Kashur K. J. Kelly W. Ketchum J. S. Kim M. King J. Klein D.-H. Koh L. Kotsiopoulou T. Kroupova V. A. Kudryavtsev V. do Lago Pimentel N. Lane J. Larkin H. Lay R. LaZur J.-Y. Li Y. Li K. Lin B. R. Littlejohn L. Liu W. C. Louis X. Lu X. Luo A. Machado P. Machado C. Mariani F. Marinho C. M. Marshall J. Marshall C. Martin-Morales S. Martynenko A. Mastbaum N. Mauri K. Mavrokoridis N. McConkey B. McCusker K. S. McFarland J. Mclaughlin A. Menegolli G. Meng O. G. Miranda A. Mogan N. Moggi E. Montagna A. Montanari C. Montanari M. Mooney A. F. Moor G. Moreno-Granados H. Da Motta C. A. Moura J. Mueller S. Mulleriababu M. Murphy D. P. Mendez D. Naples A. Navrer-Agasson M. Nebot-Guinot V. C. L. Nguyen F. J. Nicolas-Arnaldos L. Di Noto J. Nowak S. B. Oh N. Oza O. Palamara S. Palestini N. Pallat M. Pallavicini V. Pandey V. Paolone A. Papadopoulou H. B. Parkinson L. Pasqualini J. Paton L. Patrizii L. Paulucci Z. Pavlovic D. Payne L. Pelegrina-Gutierrez O. L. G. Peres G. Petrillo C. Petta V. Pia F. Pietropaolo J. Plows F. Poppi M. Pozzato M.L. Pumo G. Putnam X. Qian R. Rajagopalan A. Rappoldi G. L. Raselli P. Ratoff H. Ray M. Reggiani-Guzzo S. Repetto F. Resnati A. M. Ricci A. Roberts M. Roda A. de Roeck J. Romeo-Araujo M. Rosenberg M. Ross-Lonergan M. Rossella N. Rowe P. Roy C. Rubbia I. Safa S. Saha G. Salmoria S. Samanta A. Sanchez-Castillo P. Sanchez-Lucas A. Scaramelli D. W. Schmitz A. Schneider A. Schukraft H. Scott E. Segreto D. Senadheera S-H. Seo F. Sergiampietri M. Shaevitz P. Singh G. Sirri B. Slater J. S. Smedley J. Smith M. Soares-Nunes M. Soderberg S. Soldner-Rembold J. Spitz M. Stancari L. Stanco J. Stewart T. Strauss A. M. Szelc H. A. Tanaka M. Tenti K. Terao F. Terranova C. Thorpe V. Togo D. Torretta M. Torti F. Tortorici D. Totani M. Toups C. Touramanis R. Triozzi Y.-T. Tsai L. Tung M. Del Tutto T. Usher G. A. Valdiviesso F. Varanini N. Vardy S. Ventura M. Vicenzi C. Vignoli L. Wan R. G. Van de Water M. Weber H. Wei T. Wester A. White F.A. Wieler A. Wilkinson Z. Williams P. Wilson R. J. Wilson J. Wolfs T. Wongjirad A. Wood E. Worcester M. Worcester S. Yadav E. Yandel T. Yang L. Yates B. Yu H. Yu J. Yu B. Zamorano A. Zani A. Vazquez-Ramos J. Zennamo J. Zettlemoyer C. Zhang S. Zucchelli (ICARUS Collaboration SBND Collaboration (for the SBN Program))
This is my paper

Pith reviewed 2026-05-20 19:49 UTC · model grok-4.3

classification ⚛️ physics.ins-det hep-ex
keywords deep neural networksregion of interestliquid argon TPCneutrino detectorssignal processingSBNDICARUSionization identification
0
0 comments X

The pith

A deep neural network identifies regions of interest more effectively than traditional thresholding in liquid argon neutrino detectors.

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

The paper presents a deep neural network method, DNN ROI, for detecting regions of interest in the signal processing of liquid argon time projection chambers at the Short-Baseline Neutrino Program. It uses the full two-dimensional readout and information from multiple planes to overcome shortcomings of the standard wire-by-wire threshold approach. The authors demonstrate superior performance in identifying signals from high-energy particles and greater resistance to detector variations, whether or not they use augmented training data. This matters because better signal identification directly improves the reconstruction of neutrino interactions, supporting more precise physics results from the SBND and ICARUS detectors.

Core claim

DNN ROI addresses limitations of the traditional wire-by-wire thresholding algorithm by leveraging the full two-dimensional detector readout and cross-plane matching information. It outperforms the traditional method in both low-level ROI identification performance and high-level reconstruction metrics for high-energy cosmic and accelerator neutrino interaction products, while also being more robust against detector variations, with or without sample augmentation.

What carries the argument

DNN ROI, a deep neural network that takes advantage of the complete two-dimensional detector readout and cross-plane matching to detect regions of interest in LArTPCs.

If this is right

  • Enhanced identification of ionization charges from high-energy cosmic rays and neutrino interactions.
  • Improved metrics in the overall event reconstruction for the SBND and ICARUS detectors.
  • Increased robustness to changes in detector performance over time or between runs.
  • More reliable results in analyses of accelerator neutrino beams and cosmic ray backgrounds.

Where Pith is reading between the lines

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

  • Adapting this approach to other neutrino experiments using similar detector technology could yield comparable gains in data quality.
  • Integration into online triggering systems might reduce the volume of data needing full processing.
  • Further refinements could combine the neural network with traditional algorithms for hybrid performance.

Load-bearing premise

The samples used for training, including any augmentations, sufficiently represent the actual variations in detector response encountered during real operations of SBND and ICARUS.

What would settle it

Running the DNN ROI on real collected data from the SBND or ICARUS detectors and directly comparing its ROI identification accuracy and reconstruction quality against the traditional method on the same data set.

Figures

Figures reproduced from arXiv: 2605.18861 by A. A. Dange, A. Aduszkiewicz, A. Antonakis, A. Barnard, A. Beever, A. Bhat, A. Blake, A. Blanchet, A. Braggiotti, A. Brandt, A. Campani, A. Campos, A. Chappell, A. Chatterjee, A. Cocco, A. de Roeck, A. Ereditato, A. Ezeribe, A. Falcone, A. Fava, A. Filkins, A. F. Moor, A. Furmanski, A. Gioiosa, A. Guglielmi, A. Heggestuen, A. Hergenhan, A. Machado, A. Mastbaum, A. Menegolli, A. Mogan, A. Montanari, A. M. Ricci, A. M. Szelc, A. Navrer-Agasson, A. Papadopoulou, A. Rappoldi, A. Roberts, A. Sanchez-Castillo, A. Scaramelli, A. Schneider, A. Schukraft, A. Vazquez-Ramos, A. White, A. Wilkinson, A. Wood, A. Zani, B. Baibussinov, B. Behera, B. Bogart, B. Bottino, B. Carlson, B. Fleming, B. Howard, B. McCusker, B. R. Littlejohn, B. Slater, B. Yu, B. Zamorano, C. A. Moura, C. Andreopoulos, C. Brizzolari, C. Cuesta, C. Fan, C. Farnese, C. Gatto, C. James, C. Mariani, C. Martin-Morales, C. M. Marshall, C. Montanari, C. Petta, C. Rubbia, C. Thorpe, C. Touramanis, C. Vignoli, C. Zhang, D. Andrade Aldana, D. Brailsford, D. Caratelli, D. Carber, D. Cherdack, D. Di Ferdinando, D. Franco, D. Garcia-Gamez, D. Gibin, D.-H. Koh, D. Kalra, D. Naples, D. Payne, D. P. Mendez, D. Senadheera, D. Torretta, D. Totani, D. W. Schmitz, E. Belchior, E. Montagna, E. Segreto, E. Worcester, E. Yandel, F. Abd Alrahman, F. Akbar, F.A. Wieler, F. Boffelli, F. Castillo Fernandez, F. Cavanna, F. Drielsma, F. J. Nicolas-Arnaldos, F. Marinho, F. Pietropaolo, F. Poppi, F. Resnati, F. Sergiampietri, F. Terranova, F. Tortorici, F. Varanini, G. A. Valdiviesso, G. Cerati, G. Fricano, G. Gurung, G. Karagiorgi, G. L. Raselli, G. Meng, G. Moreno-Granados, G. Petrillo, G. Putnam, G. Salmoria, G. Sirri, H. A. Tanaka, H. B. Parkinson, H. Carranza, H. Chen, H. da Motta, H. Hausner, H. Lay, H. Ray, H. S. Budd, H. Scott, H. Wei, H. Yu, I. Caro Terrazas, I. Furic, I. Gil-Botella, I. Ingratta, I. Safa, J. Asaadi, J. Bateman, J. Berger, J. Dyer, J. J. Evans, J. Klein, J. Larkin, J. Marshall, J. Mclaughlin, J. Mueller, J. Nowak, J. Paton, J. Plows, J. Romeo-Araujo, J. S. Kim, J. Smith, J. Spitz, J. S. Smedley, J. Stewart, J. Wolfs, J.-Y. Li, J. Yu, J. Zennamo, J. Zettlemoyer, K. Duffy, K. Hassinin, K. J. Kelly, K. Lin, K. Mavrokoridis, K. S. McFarland, K. Terao, L. Aliaga-Soplin, L. Camilleri, L. Cooper-Troendle, L. Di Noto, L. Hagaman, L. Kashur, L. Kotsiopoulou, L. Liu, L. Pasqualini, L. Patrizii, L. Paulucci, L. Pelegrina-Gutierrez, L. Stanco, L. Tung, L. Wan, L. Yates, M. Artero Pons, M. B. Brunetti, M. Betancourt, M. Bishai, M. Bonesini, M. Cicerchia, M. Dall'Olio, M. Del Tutto, M. Diwan, M. Dubnowski, M. F. Carneiro, M. F. Cicala, M. Hernandez-Morquecho, M. Jung, M. King, M.L. Pumo, M. Mooney, M. Murphy, M. Nebot-Guinot, M. Pallavicini, M. Pozzato, M. R. Convery, M. Reggiani-Guzzo, M. Roda, M. Rosenberg, M. Rossella, M. Ross-Lonergan, M. Shaevitz, M. S. Ismail, M. Soares-Nunes, M. Soderberg, M. Stancari, M. Tenti, M. Torti, M. Toups, M. Vicenzi, M. Weber, M. Worcester, N. Abrego-Martinez, N. Chithirasreemadam, N. Gallice, N. Lane, N. Mauri, N. McConkey, N. Moggi, N. Oza, N. Pallat, N. Rowe, N. Vardy, O. Dalager, O. G. Miranda, O. L. G. Peres, O. Palamara, P. Abratenko, P. Green, P. Hamilton, P. Holanda, P. Machado, P. Ratoff, P. Roy, P. Sanchez-Lucas, P. Singh, P. Wilson, R. Acciarri, R. Alvarez-Garrote, R. Benocci, R. Castillo, R. Coackley, R. Darby, R. G. Van de Water, R. Howell, R. J. Wilson, R. LaZur, R. Rajagopalan, R. S. Jones, R. Triozzi, S. Baena, S. Balasubramanian, S. Bertolucci, SBND Collaboration (for the SBN Program)), S. B. Oh, S. Brickner, S. Centro, S. Cherubini, S. Chung, S. Copello, S. Di Domizio, S. Dolan, S. Dominguez-Vidales, S. Donati, S. Dytman, S. Gao, S. Gardiner, S. Gollapinni, S-H. Seo, S. J. Brice, S. Kr Das, S. Martynenko, S. Mulleriababu, S. Palestini, S. Repetto, S. Saha, S. Samanta, S. Soldner-Rembold, S. Ventura, S. Yadav, S. Zucchelli (ICARUS Collaboration, T. Boone, T. E. Coan, T. Junk, T. Kroupova, T. Strauss, T. Usher, T. Wester, T. Wongjirad, T. Yang, V. A. Kudryavtsev, V. Basque, V. Bellini, V. Brio, V. C. L. Nguyen, V. do Lago Pimentel, V. Pandey, V. Paolone, V. Pia, V. Togo, W. C. Griffith, W. C. Louis, W. F. Badgett, W. Foreman, W. Gu, W. Jang, W. Ketchum, X. Lu, X. Luo, X. Qian, Y. Dabburi, Y.-J. Jwa, Y. Li, Y.-T. Tsai, Z. Djurcic, Z. Hulcher, Z. Pavlovic, Z. Williams.

Figure 1
Figure 1. Figure 1: FIG. 1. Structure of the DNN ROI network applied in SBND and ICARUS. The network applies a U-ResNet architecture [9, 10], [PITH_FULL_IMAGE:figures/full_fig_p004_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: FIG. 2. Example deconvolved waveform on the front in [PITH_FULL_IMAGE:figures/full_fig_p006_2.png] view at source ↗
Figure 3
Figure 3. Figure 3: FIG. 3. Variation in the single electron response across the [PITH_FULL_IMAGE:figures/full_fig_p007_3.png] view at source ↗
Figure 4
Figure 4. Figure 4: FIG. 4. Demonstration of image augmentations used to simulate detector variations on the middle induction plane in ICARUS. [PITH_FULL_IMAGE:figures/full_fig_p008_4.png] view at source ↗
Figure 5
Figure 5. Figure 5: 1 N(µ, σ2 ) denotes a normal distribution with mean µ and stan￾dard deviation σ IV. NETWORK OPTIMIZATION AND TRAINING For application to SBN detectors, computational effi￾ciency is a critical requirement. The trained networks need to be able to run on CPUs with reasonable memory and time requirements to be applied in data processing. To satisfy these conditions, we adopted a strategy of reducing the input … view at source ↗
Figure 5
Figure 5. Figure 5: FIG. 5. Demonstration of image augmentations used to simulate detector variations on the middle induction plane in SBND. [PITH_FULL_IMAGE:figures/full_fig_p009_5.png] view at source ↗
Figure 6
Figure 6. Figure 6: FIG. 6. Validation loss for three different optimized model [PITH_FULL_IMAGE:figures/full_fig_p009_6.png] view at source ↗
Figure 8
Figure 8. Figure 8: FIG. 8. Comparison of the performance of traditional and [PITH_FULL_IMAGE:figures/full_fig_p011_8.png] view at source ↗
Figure 7
Figure 7. Figure 7: FIG. 7. Comparison of performance of traditional and [PITH_FULL_IMAGE:figures/full_fig_p011_7.png] view at source ↗
Figure 9
Figure 9. Figure 9: FIG. 9. ROI Efficiency [PITH_FULL_IMAGE:figures/full_fig_p011_9.png] view at source ↗
Figure 11
Figure 11. Figure 11: FIG. 11. Pixelwise ROI identification efficiency [PITH_FULL_IMAGE:figures/full_fig_p012_11.png] view at source ↗
Figure 13
Figure 13. Figure 13: FIG. 13. The DNN prediction score for true signal and non [PITH_FULL_IMAGE:figures/full_fig_p013_13.png] view at source ↗
Figure 14
Figure 14. Figure 14: FIG. 14. Comparison of ROI efficiency [PITH_FULL_IMAGE:figures/full_fig_p013_14.png] view at source ↗
Figure 16
Figure 16. Figure 16: FIG. 16. The fractional offset between the total measured [PITH_FULL_IMAGE:figures/full_fig_p014_16.png] view at source ↗
read the original abstract

We present a deep neural net-based region of interest detection method (DNN ROI) for signal processing in the liquid argon time projection chambers of the Short-Baseline Neutrino (SBN) Program, SBND and ICARUS. DNN ROI addresses limitations of the traditional wire-by-wire thresholding algorithm by leveraging the full two-dimensional detector readout and cross-plane matching information. To account for detector performance variations, we explore training with augmented samples. We find that DNN ROI outperforms the traditional method in both low-level ROI identification performance and high-level reconstruction metrics for high-energy cosmic and accelerator neutrino interaction products, while also being more robust against detector variations, with or without sample augmentation.

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 / 1 minor

Summary. The manuscript introduces a deep neural network method (DNN ROI) for region-of-interest detection in the signal processing chain of liquid argon time projection chambers in the SBN Program detectors (SBND and ICARUS). It claims that DNN ROI outperforms the conventional wire-by-wire thresholding algorithm on both low-level ROI identification metrics and downstream high-level reconstruction quantities for high-energy cosmic-ray and accelerator neutrino interactions, while also demonstrating greater robustness to detector variations whether or not the training set includes augmented samples that simulate performance fluctuations.

Significance. If the reported gains and robustness hold under realistic conditions, the approach could reduce systematic uncertainties in ionization charge identification and improve event reconstruction fidelity in LArTPC-based neutrino experiments, with direct relevance to the physics goals of the SBN program.

major comments (2)
  1. The central robustness claim—that DNN ROI remains superior “with or without sample augmentation” and is “more robust against detector variations”—rests on the assumption that the augmentation procedure faithfully reproduces the joint statistics of real effects (plane-to-plane correlated noise, gain drifts, wire response variations, time-dependent pedestal shifts). No quantitative closure test comparing augmented-sample distributions to real calibration-run statistics is presented; this is load-bearing for extrapolating the observed margin to live SBND/ICARUS operation.
  2. Comparative performance results for low-level ROI identification and high-level reconstruction metrics are reported without error bars, without explicit description of the validation-split protocol, and without quantitative thresholds defining “outperformance.” These omissions leave the strength of the headline claims only moderately supported.
minor comments (1)
  1. Figure captions and axis labels should explicitly state whether the plotted distributions are from augmented or unaugmented test sets.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for their detailed and constructive comments on our manuscript. We address each of the major comments below and indicate the revisions we plan to make.

read point-by-point responses
  1. Referee: The central robustness claim—that DNN ROI remains superior “with or without sample augmentation” and is “more robust against detector variations”—rests on the assumption that the augmentation procedure faithfully reproduces the joint statistics of real effects (plane-to-plane correlated noise, gain drifts, wire response variations, time-dependent pedestal shifts). No quantitative closure test comparing augmented-sample distributions to real calibration-run statistics is presented; this is load-bearing for extrapolating the observed margin to live SBND/ICARUS operation.

    Authors: We recognize the importance of validating the augmentation procedure against real data. While the augmentation is constructed from individually measured detector effects, a direct closure test on joint distributions was not included in the original submission. In the revised manuscript, we will add a quantitative comparison using calibration data from the SBND and ICARUS detectors to assess how well the augmented samples reproduce the observed statistics of noise, gain, and pedestal variations. revision: yes

  2. Referee: Comparative performance results for low-level ROI identification and high-level reconstruction metrics are reported without error bars, without explicit description of the validation-split protocol, and without quantitative thresholds defining “outperformance.” These omissions leave the strength of the headline claims only moderately supported.

    Authors: We agree that including error bars, detailing the validation protocol, and specifying quantitative outperformance criteria will better support our claims. We will revise the manuscript to include statistical uncertainties on all performance metrics, describe the train/validation/test split and any cross-validation procedure used, and provide explicit thresholds or significance levels for claiming outperformance. revision: yes

Circularity Check

0 steps flagged

No significant circularity: empirical performance comparison is self-contained

full rationale

The paper reports an empirical study comparing a DNN-based ROI detection method to a traditional wire-by-wire thresholding algorithm on liquid argon TPC data from the SBN program. Performance is quantified via standard metrics (e.g., identification efficiency, reconstruction quality) evaluated on held-out test sets and augmented samples that simulate detector variations. No equations, first-principles derivations, or predictions are presented that reduce by construction to fitted parameters or inputs defined from the same evaluation data. The robustness claims rest on the external assumption that the augmentation procedure adequately models real detector effects, but this is a methodological limitation rather than a logical circularity in which a claimed result is definitionally equivalent to its training inputs. No self-citation chains, uniqueness theorems, or ansatzes are invoked to force the central result. The derivation chain is therefore independent and externally falsifiable through direct comparison on real or simulated data.

Axiom & Free-Parameter Ledger

1 free parameters · 2 axioms · 0 invented entities

The central performance and robustness claims rest on the assumption that the training distribution (including augmentations) matches real detector behavior and that standard supervised learning generalizes from the chosen metrics.

free parameters (1)
  • DNN architecture and training hyperparameters
    Network depth, learning rate, augmentation parameters, and loss weights are selected or optimized during development and affect the reported gains.
axioms (2)
  • domain assumption Detector signals can be usefully represented as 2D images for convolutional processing.
    Invoked when the method shifts from wire-by-wire to full 2D readout.
  • ad hoc to paper Augmented samples sufficiently span the space of detector variations.
    Required for the robustness claim to hold beyond the training distribution.

pith-pipeline@v0.9.0 · 7459 in / 1439 out tokens · 41079 ms · 2026-05-20T19:49:33.464196+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

47 extracted references · 47 canonical work pages · 13 internal anchors

  1. [1]

    ROI Filter Output: This frame consists of the de- convolved signal on the wire plane, using the ROI filter (Eq. 5)

  2. [2]

    It identifies channels where activity in one plane coin- cides with activity in at least one other plane within a common time window on an overlapping wire

    Two-Plane (MP2) Coincidence: This frame en- codes geometric constraints across wire planes. It identifies channels where activity in one plane coin- cides with activity in at least one other plane within a common time window on an overlapping wire. This information helps the network disambiguate true physics signals from noise and artifacts that are unlik...

  3. [3]

    Nominal” sample, detector simulation pa- rameters were fixed at their nominal values. In the “Om- niDetector

    Three-Plane (MP3) Coincidence: Similar to MP2, this frame highlights channels where signals are simultaneously present (in time) across all three wire planes, providing the strongest geometric con- straint and a robust handle on true charge deposi- tions. Signals over five times the noise RMS on the plane of interest, and signals over three times the nois...

  4. [4]

    R. Acciarriet al.(MicroBooNE, LAr1-ND, ICARUS- WA104), A Proposal for a Three Detector Short- Baseline Neutrino Oscillation Program in the Fermilab Booster Neutrino Beam, arxiv (2015), arXiv:1503.01520 [physics.ins-det]

  5. [5]

    P. A. Machado, O. Palamara, and D. W. Schmitz, The Short-Baseline Neutrino Program at Fermilab, Ann. Rev. Nucl. Part. Sci.69, 363 (2019), arXiv:1903.04608 [hep- ex]

  6. [6]

    Abratenkoet al.(ICARUS), ICARUS at the Fermilab Short-Baseline Neutrino program: initial operation, Eur

    P. Abratenkoet al.(ICARUS), ICARUS at the Fermilab Short-Baseline Neutrino program: initial operation, Eur. Phys. J. C83, 467 (2023), arXiv:2301.08634 [hep-ex]

  7. [7]

    Acciarriet al.(MicroBooNE), Design and Construc- tion of the MicroBooNE Detector, JINST12(02), P02017, arXiv:1612.05824 [physics.ins-det]

    R. Acciarriet al.(MicroBooNE), Design and Construc- tion of the MicroBooNE Detector, JINST12(02), P02017, arXiv:1612.05824 [physics.ins-det]

  8. [8]

    A. A. Aguilar-Arevaloet al.(MiniBooNE), The Neutrino Flux Prediction at MiniBooNE, Phys. Rev. D79, 072002 (2009), arXiv:0806.1449 [hep-ex]

  9. [9]

    The NuMI Neutrino Beam

    P. Adamsonet al., The NuMI Neutrino Beam, Nucl. Instrum. Meth. A806, 279 (2016), arXiv:1507.06690 [physics.acc-ph]

  10. [10]

    Adamset al.(MicroBooNE), Ionization electron signal processing in single phase LArTPCs

    C. Adamset al.(MicroBooNE), Ionization electron signal processing in single phase LArTPCs. Part I. Algorithm Description and quantitative evaluation with MicroBooNE simulation, JINST13(07), P07006, arXiv:1802.08709 [physics.ins-det]

  11. [11]

    Yuet al., Augmented signal processing in Liquid Argon Time Projection Chambers with a deep neu- ral network, JINST16(01), P01036, arXiv:2007.12743 [physics.ins-det]

    H. Yuet al., Augmented signal processing in Liquid Argon Time Projection Chambers with a deep neu- ral network, JINST16(01), P01036, arXiv:2007.12743 [physics.ins-det]

  12. [12]

    U-Net: Convolutional Networks for Biomedical Image Segmentation

    O. Ronneberger, P. Fischer, and T. Brox, U-net: Con- volutional networks for biomedical image segmentation, CoRRabs/1505.04597(2015), 1505.04597

  13. [14]

    Baldi, J

    T. Baldi, J. Campos, O. Weng, C. Geniesse, N. Tran, R. Kastner, and A. Biondi, Loss Landscape Anal- ysis for Reliable Quantized ML Models for Scien- tific Sensing, arXiv e-prints , arXiv:2502.08355 (2025), arXiv:2502.08355 [cs.LG]

  14. [15]

    Kasieczka and D

    G. Kasieczka and D. Shih, Robust Jet Classifiers through Distance Correlation, Phys. Rev. Lett.125, 122001 (2020), arXiv:2001.05310 [hep-ph]

  15. [16]

    Ghosh, B

    A. Ghosh, B. Nachman, and D. Whiteson, Uncertainty- aware machine learning for high energy physics, Phys. Rev. D104, 056026 (2021), arXiv:2105.08742 [physics.data-an]

  16. [17]

    Wilkinson, R

    A. Wilkinson, R. Radev, and S. Alonso-Monsalve, Con- trastive learning for robust representations of neutrino data, Phys. Rev. D111, 092011 (2025), arXiv:2502.07724 [hep-ex]

  17. [18]

    A Convolutional Neural Network Neutrino Event Classifier

    A. Aurisano, A. Radovic, D. Rocco, A. Himmel, M. D. Messier, E. Niner, G. Pawloski, F. Psihas, A. Sousa, and P. Vahle, A Convolutional Neural Net- work Neutrino Event Classifier, JINST11(09), P09001, arXiv:1604.01444 [hep-ex]

  18. [19]

    A. M. Sirunyanet al.(CMS), Identification of heavy- flavour jets with the CMS detector in pp collisions at 13 TeV, JINST13(05), P05011, arXiv:1712.07158 [physics.ins-det]

  19. [20]

    Efficient Antihydrogen Detection in Antimatter Physics by Deep Learning

    P. Sadowski, B. Radics, Ananya, Y. Yamazaki, and P. Baldi, Efficient antihydrogen detection in antimat- ter physics by deep learning, J. Phys. Comm.1, 025001 (2017), arXiv:1706.01826 [physics.ins-det]

  20. [21]

    Adamset al.(MicroBooNE), Deep neural network for pixel-level electromagnetic particle identification in the MicroBooNE liquid argon time projection chamber, Phys

    C. Adamset al.(MicroBooNE), Deep neural network for pixel-level electromagnetic particle identification in the MicroBooNE liquid argon time projection chamber, Phys. Rev. D99, 092001 (2019), arXiv:1808.07269 [hep- ex]

  21. [22]

    Decorrelated Jet Substructure Tagging using Adversarial Neural Networks

    C. Shimmin, P. Sadowski, P. Baldi, E. Weik, D. White- son, E. Goul, and A. Søgaard, Decorrelated Jet Substruc- ture Tagging using Adversarial Neural Networks, Phys. Rev. D96, 074034 (2017), arXiv:1703.03507 [hep-ex]

  22. [23]

    Improved Energy Reconstruction in NOvA with Regression Convolutional Neural Networks

    P. Baldi, J. Bian, L. Hertel, and L. Li, Improved En- ergy Reconstruction in NOvA with Regression Convolu- tional Neural Networks, Phys. Rev. D99, 012011 (2019), arXiv:1811.04557 [physics.ins-det]

  23. [25]

    Radovic, M

    A. Radovic, M. Williams, D. Rousseau, M. Kagan, D. Bonacorsi, A. Himmel, A. Aurisano, K. Terao, and T. Wongjirad, Machine learning at the energy and inten- sity frontiers of particle physics, Nature560, 41 (2018)

  24. [26]

    R. Acciarriet al.(ArgoNeuT), A deep-learning based raw waveform region-of-interest finder for the liquid ar- gon time projection chamber, JINST17(01), P01018, arXiv:2103.06391 [physics.ins-det]

  25. [27]

    Uboldi, D

    L. Uboldi, D. Ruth, M. Andrews, M. H. L. S. Wang, H. J. Wenzel, W. Wu, and T. Yang, Extracting low energy signals from raw LArTPC waveforms using deep learning techniques — A proof of concept, Nucl. Instrum. Meth. A 1028, 166371 (2022), arXiv:2106.09911 [physics.ins-det]

  26. [28]

    Abiet al.(DUNE), Deep Underground Neutrino Experiment (DUNE), Far Detector Technical Design Report, Volume II: DUNE Physics, arxiv (2020), arXiv:2002.03005 [hep-ex]

    B. Abiet al.(DUNE), Deep Underground Neutrino Experiment (DUNE), Far Detector Technical Design Report, Volume II: DUNE Physics, arxiv (2020), arXiv:2002.03005 [hep-ex]

  27. [29]

    SBND (SBND), Performance and Description of the Short-Baseline Near Detector (SBND), (2025)

  28. [30]

    P. Abratenkoet al.(ICARUS), Calibration and simu- lation of ionization signal and electronics noise in the ICARUS liquid argon time projection chamber, JINST 20(01), P01032, arXiv:2407.11925 [hep-ex]

  29. [31]

    Wiener,Extrapolation, Interpolation and Smoothing of Stationary Time Series(M.I.T

    N. Wiener,Extrapolation, Interpolation and Smoothing of Stationary Time Series(M.I.T. Press, Cambridge, Mass., 1949)

  30. [32]

    E. L. Snider and G. Petrillo, LArSoft: Toolkit for Sim- ulation, Reconstruction and Analysis of Liquid Argon TPC Neutrino Detectors, J. Phys. Conf. Ser.898, 042057 (2017)

  31. [33]

    The GENIE Neutrino Monte Carlo Generator

    C. Andreopouloset al., The GENIE Neutrino Monte Carlo Generator, Nucl. Instrum. Meth.A614, 87 (2010), arXiv:0905.2517 [hep-ph]

  32. [34]

    Alvarez-Rusoet al.(GENIE), Recent highlights from GENIE v3, Eur

    L. Alvarez-Rusoet al.(GENIE), Recent highlights from GENIE v3, Eur. Phys. J. ST230, 4449 (2021), arXiv:2106.09381 [hep-ph]

  33. [35]

    D. Heck, J. Knapp, J. N. Capdevielle, G. Schatz, and T. Thouw, CORSIKA: A Monte Carlo code to simulate extensive air showers (1998)

  34. [36]

    S. Agostinelliet al., Geant4—a simulation toolkit, Nu- clear Instruments and Methods in Physics Research Sec- tion A: Accelerators, Spectrometers, Detectors and As- sociated Equipment506, 250 (2003)

  35. [37]

    P. Abratenkoet al.(ICARUS), Angular dependent mea- surement of electron-ion recombination in liquid argon for ionization calorimetry in the ICARUS liquid ar- gon time projection chamber, JINST20(01), P01033, arXiv:2407.12969 [physics.ins-det]

  36. [38]

    Bunemann, T

    O. Bunemann, T. Cranshaw, and J. Harvey, Design of grid ionization chambers, Canadian Journal of Research 27, 191 (2011)

  37. [39]

    G¨ opfert, F

    A. G¨ opfert, F. josef Hambsch, and H. Bax, A twin ioniza- tion chamber setup as detector for light charged particles with energies around 1mev applied to the reaction, Nu- clear Instruments & Methods in Physics Research Section A-accelerators Spectrometers Detectors and Associated Equipment441, 438 (2000)

  38. [40]

    Bevilacqua, A

    R. Bevilacqua, A. G¨ o¨ ok, F. J. Hambsch, N. Jovanˇ cevi´ c, and M. Vidali, A procedure for the characterization of electron transmission through Frisch grids, Nucl. In- strum. Meth. A770, 64 (2015)

  39. [41]

    3D Semantic Segmentation with Submanifold Sparse Convolutional Networks

    B. Graham, M. Engelcke, and L. van der Maaten, 3d semantic segmentation with submanifold sparse con- volutional networks, CoRRabs/1711.10275(2017), 1711.10275

  40. [42]

    Drielsma, K

    F. Drielsma, K. Terao, L. Domin´ e, and D. H. Koh, Scalable, end-to-end, deep-learning-based data recon- struction chain for particle imaging detectors (2021), arXiv:2102.01033 [hep-ex]

  41. [43]

    Z. Zhou, M. M. R. Siddiquee, N. Tajbakhsh, and J. Liang, Unet++: A nested u-net architecture for medi- cal image segmentation, CoRRabs/1807.10165(2018), 1807.10165

  42. [44]

    L. R. Dice, Measures of the amount of ecologic associa- tion between species, Ecology26, 297 (1945)

  43. [45]

    T. Sørensen, A method of establishing groups of equal amplitude in plant sociology based on similarity of species content and its application to analyses of the veg- etation on Danish commons, Kongelige Danske Vidensk- abernes Selskab, Biologiske Skrifter5, 1 (1948)

  44. [46]

    A. P. Zijdenbos, B. M. Dawant, R. A. Margolin, and A. C. 17 Palmer, Morphometric analysis of white matter lesions in MR images: Method and validation, IEEE Transactions on Medical Imaging13, 716 (1994)

  45. [47]

    K. He, X. Zhang, S. Ren, and J. Sun, Deep residual learning for image recognition (2015), arXiv:1512.03385 [cs.CV]

  46. [48]

    D. P. Kingma and J. Ba, Adam: A method for stochastic optimization (2017), arXiv:1412.6980 [cs.LG]

  47. [49]

    Paszke, S

    A. Paszke, S. Gross, F. Massa, J. Lerer, Soumith Chin- tala, G. DeVito, Z. Lin, L. Antiga, A. Rizan, T. Penny, et al., Pytorch: An imperative style, high-performance deep learning library, Advances in Neural Information Processing Systems32(2019)