pith. sign in

arxiv: 2606.29381 · v1 · pith:MTE4WKRBnew · submitted 2026-06-28 · 🌌 astro-ph.HE · hep-ex

Search for Diffuse Supernova Neutrino Background in the Full KamLAND Dataset with Neural-Network-Based Event Classification

D. Chernyak , T. Eda , M. Eizuka , R. Endo , A. Gando , Y. Gando , T. Hachiya , F. Haneishi
show 58 more authors
K. Hata T. Hirai K. Hosokawa K. Ichimura H. Ikeda K. Inoue K. Ishidoshiro Y. Kamei N. Kawada Y. Kishimoto M. Koga K. Mikami H. Miyake K. Mizukoshi D. Morita K. Nakamura R. Nakamura J. Nakane Y. Nakano N. Obata Y. Ota K. Saito I. Sakaki I. Shimizu J. Shirai A. Suzuki K. Tachibana A. Takeuchi K. Tamae H. Watanabe Z. Xu S. Yoshida S. Umehara K. Fushimi K. Kotera Y. Urano S. Kurosawa H. Ozaki B. E. Berger B. K. Fujikawa J. G. Learned J. Maricic Z. Li L. A. Winslow Y. Efremenko H. J. Karwowski D. M. Markoff W. Tornow S. Dell'Oro T. O'Donnell J. A. Detwiler S. Enomoto M. P. Decowski K. M. Weerman C. Grant O. Penek H. Song A. Li
This is my paper

Pith reviewed 2026-06-30 02:13 UTC · model grok-4.3

classification 🌌 astro-ph.HE hep-ex
keywords diffuse supernova neutrino backgroundKamLANDneural networkevent classificationinverse beta decayupper limitsneutrino fluxliquid scintillator
0
0 comments X

The pith

KamLAND observes seven events consistent with background and sets 90% CL upper limits of 38-43 per square centimeter per second on the diffuse supernova neutrino background flux.

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

The paper conducts a search for the diffuse supernova neutrino background by detecting electron antineutrinos via inverse beta decay in the KamLAND liquid scintillator detector. It employs a deep neural network to classify events and suppress neutron backgrounds across exposures of roughly 9 kton-years in the 8.3-30.8 MeV range. Only seven candidates are found, matching the expected background of 16.2 plus or minus 9.4 events. Spectral analysis of energy and radial distributions shows no excess signal. This leads to new upper limits on the DSNB flux and demonstrates the neural network method for background reduction.

Core claim

Using neural-network event classification on the full KamLAND dataset, seven inverse-beta-decay candidates are observed in the 8.3-30.8 MeV range against a background expectation of 16.2 ± 9.4 events; a fit to the energy and radial distributions finds no significant DSNB excess, yielding 90% confidence-level flux upper limits of 38-43 per square centimeter per second depending on the model, plus model-independent antineutrino flux limits that are among the strongest below 13.3 MeV.

What carries the argument

Deep neural network for inverse-beta-decay event classification that suppresses neutron-associated backgrounds while preserving signal efficiency.

If this is right

  • The DSNB flux lies below the reported 90% CL limits for each tested model.
  • Model-independent electron-antineutrino flux limits are among the most stringent below 13.3 MeV.
  • Neural-network classification can suppress neutron backgrounds in other liquid-scintillator detectors.
  • The method enables tighter DSNB searches with future higher-exposure data.

Where Pith is reading between the lines

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

  • Combining these limits with data from water-Cherenkov detectors could further constrain the DSNB spectrum shape.
  • The neural-network approach may extend to other rare-event searches where neutron backgrounds dominate.
  • Updated supernova rate models or neutrino oscillation parameters could be tested against these bounds in future analyses.

Load-bearing premise

The neural network identifies inverse beta decay candidates without bias or unaccounted efficiency loss, and the background prediction of 16.2 plus or minus 9.4 events captures all relevant systematic uncertainties.

What would settle it

A statistically significant excess of events whose energy and radial distributions match a DSNB spectrum above the quoted background would contradict the no-signal result.

Figures

Figures reproduced from arXiv: 2606.29381 by A. Gando, A. Li, A. Suzuki, A. Takeuchi, B. E. Berger, B. K. Fujikawa, C. Grant, D. Chernyak, D. M. Markoff, D. Morita, F. Haneishi, H. Ikeda, H. J. Karwowski, H. Miyake, H. Ozaki, H. Song, H. Watanabe, I. Sakaki, I. Shimizu, J. A. Detwiler, J. G. Learned, J. Maricic, J. Nakane, J. Shirai, K. Fushimi, K. Hata, K. Hosokawa, K. Ichimura, K. Inoue, K. Ishidoshiro, K. Kotera, K. Mikami, K. Mizukoshi, K. M. Weerman, K. Nakamura, K. Saito, K. Tachibana, K. Tamae, L. A. Winslow, M. Eizuka, M. Koga, M. P. Decowski, N. Kawada, N. Obata, O. Penek, R. Endo, R. Nakamura, S. Dell'Oro, S. Enomoto, S. Kurosawa, S. Umehara, S. Yoshida, T. Eda, T. Hachiya, T. Hirai, T. O'Donnell, W. Tornow, Y. Efremenko, Y. Gando, Y. Kamei, Y. Kishimoto, Y. Nakano, Y. Ota, Y. Urano, Z. Li, Z. Xu.

Figure 1
Figure 1. Figure 1: KamNet score distribution for the IBD signal and atmospheric-neutrino NC background. This figure corresponds to the period before purification [PITH_FULL_IMAGE:figures/full_fig_p006_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: Schematic illustration of the proxy-sample strategy used to evaluate the systematic uncertainty associated with data–simulation differences in the KamNet selection. Simulated IBD events and atmospheric NC events are used to define the signal-like and NC-like classes, respectively. The event-discrimination arrow represents the application of the trained KamNet classifier, not the training procedure itself. … view at source ↗
Figure 3
Figure 3. Figure 3: Event distributions for all events before and after the KamNet selection. Upper left: prompt￾energy distribution; upper right: delayed-energy distribution; lower left: distance between the prompt and delayed events; lower right: time difference between the prompt and delayed events. The blue histograms correspond to events before the KamNet selection, while the red histograms represent events after the Kam… view at source ↗
Figure 4
Figure 4. Figure 4: Position distributions of the final DSNB candidates. The left panel shows prompt event positions while the right panel shows delayed event positions. Blue curves indicate, respectively, the inner-balloon region, whose interior is excluded from the analysis, the 550 cm radius, and the 650 cm radius. expected events, indicating no significant tension with the prediction [PITH_FULL_IMAGE:figures/full_fig_p01… view at source ↗
Figure 5
Figure 5. Figure 5: Two-dimensional scan of the number of DSNB and atmospheric-neutrino NC events. The DSNB model of Horiuchi et al. (2009) is adopted. Color contours indicate the 1σ (red), 90% (orange), 2σ (green), and 3σ (blue) allowed regions. The best-fit numbers of DSNB and NC events are both 0.0 (black circle). The horizontally hatched region denotes the expected number of NC events with its 1σ uncertainty, while the ve… view at source ↗
Figure 6
Figure 6. Figure 6: Prompt energy spectrum and radial distribution of the best-fit backgrounds and the DSNB signal at the 90% C.L. upper limit. The DSNB model of Horiuchi et al. (2009) is adopted. All histograms are stacked [PITH_FULL_IMAGE:figures/full_fig_p015_6.png] view at source ↗
Figure 7
Figure 7. Figure 7: The 90% C.L. upper limits on the model-independent ¯νe flux. The red points represent the limits obtained in this work, while the lighter red points indicate the limits from the previous KamLAND result (Abe et al. 2022). Other points show the limits reported by Borexino (Agostini et al. 2021) and Super￾Kamiokande during the SK-I, II, III (Bays et al. 2012), SK-IV (Abe et al. 2021), and SK-VI (Harada et al.… view at source ↗
read the original abstract

We report a search for the diffuse supernova neutrino background (DSNB) with the KamLAND detector, targeting electron antineutrinos via inverse beta decay in the neutrino energy range of 8.3 to 30.8 MeV. Using liquid-scintillator exposures of 9.02 kton-year for 8.3 to 9.3 MeV and 9.42 kton-year for 9.3 to 30.8 MeV, we observe seven candidate events after applying a new deep-neural-network-based event classification technique. This result is consistent with the background-only expectation of 16.2 plus or minus 9.4 events, including systematic uncertainties associated with the neural-network selection. A spectral analysis of the energy and radial distributions finds no significant excess attributable to the DSNB. We therefore set 90 percent confidence-level upper limits on the DSNB flux of 38 to 43 per square centimeter per second, depending on the assumed DSNB model. We also derive model-independent 90 percent confidence-level upper limits on the electron-antineutrino flux, obtaining some of the most stringent constraints below 13.3 MeV. Beyond the DSNB search itself, this work demonstrates neural-network-based event classification as a promising approach for suppressing neutron-associated backgrounds in liquid-scintillator neutrino detectors.

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

1 major / 0 minor

Summary. The manuscript reports a search for the diffuse supernova neutrino background (DSNB) via inverse beta decay in KamLAND, using 9.02–9.42 kton-year exposures and a new deep neural network for event classification. Seven candidate events are observed in the 8.3–30.8 MeV range, consistent with a background expectation of 16.2 ± 9.4 events that incorporates NN-related systematics. A spectral fit to energy and radial distributions finds no significant excess, yielding 90% CL upper limits of 38–43 cm⁻² s⁻¹ on the DSNB flux (model-dependent) and model-independent limits below 13.3 MeV. The work also positions NN classification as a tool for neutron-background suppression in liquid-scintillator detectors.

Significance. If the NN performance and systematic treatment are robust, the result supplies competitive DSNB constraints and illustrates a practical advance in background rejection for future experiments. The large quoted background uncertainty (~58%) already tempers the strength of the “no excess” statement, so the limits are driven more by the observed count and the spectral shape than by precise background subtraction.

major comments (1)
  1. [Abstract and neural-network selection description] The central claim that the background expectation of 16.2 ± 9.4 events fully captures all NN-related systematics (Abstract) is load-bearing for both the consistency statement and the derived limits. Without reported validation (e.g., signal and background efficiency versus energy or radius, or data-MC comparisons of the NN score distribution), it remains possible that an unaccounted energy- or radius-dependent mismatch between the NN response to IBD signal and the backgrounds used in training propagates into the spectral fit, weakening the “no excess” conclusion and the quoted 38–43 cm⁻² s⁻¹ bounds.

Simulated Author's Rebuttal

1 responses · 0 unresolved

We thank the referee for their thorough review and for highlighting the importance of validating the neural-network (NN) systematic treatment. We address the single major comment below. Where the concern identifies a presentational gap, we have revised the manuscript to include the requested validation material; the underlying analysis and quoted limits are unchanged.

read point-by-point responses

  1. Referee: The central claim that the background expectation of 16.2 ± 9.4 events fully captures all NN-related systematics (Abstract) is load-bearing for both the consistency statement and the derived limits. Without reported validation (e.g., signal and background efficiency versus energy or radius, or data-MC comparisons of the NN score distribution), it remains possible that an unaccounted energy- or radius-dependent mismatch between the NN response to IBD signal and the backgrounds used in training propagates into the spectral fit, weakening the “no excess” conclusion and the quoted 38–43 cm⁻² s⁻¹ bounds.

    Authors: We agree that explicit validation strengthens the claim that the quoted uncertainty fully encompasses NN-related effects. The ±9.4 uncertainty was obtained by propagating variations in NN training samples, cut thresholds, and input-feature modeling through the full analysis chain (see Section 4.3 and Appendix B). Nevertheless, to directly address the referee’s request we have added two new figures in the revised manuscript: (i) signal and background selection efficiencies versus reconstructed energy and radial position, and (ii) data–MC comparisons of the NN output score in sideband regions. These additions confirm that residual energy- or radius-dependent discrepancies lie within the already-assigned systematic envelope. Because the spectral fit is performed on the observed event count and shape, and the background uncertainty is already large (~58 %), the 90 % CL limits of 38–43 cm⁻² s⁻¹ remain unchanged. We have also updated the abstract to reference the new validation material. revision: yes

Circularity Check

0 steps flagged

No significant circularity in derivation chain

full rationale

The paper performs an experimental search by counting candidate events after NN classification and comparing the observed number (7) and spectral distributions directly to an independently modeled background expectation (16.2 ± 9.4) that already folds in NN-related systematics. Upper limits are obtained from a standard frequentist spectral fit finding no excess; no equations, fitted parameters, or self-citations reduce the reported limits to the input data by construction. The analysis rests on external data comparison rather than any self-referential definition or ansatz.

Axiom & Free-Parameter Ledger

1 free parameters · 1 axioms · 0 invented entities

The analysis rests on standard assumptions of inverse beta decay kinematics and background composition in liquid scintillator; the neural network itself introduces hyperparameters whose effect on efficiency and purity is not quantified in the abstract.

free parameters (1)
  • neural-network decision threshold
    A classification threshold must be chosen to define the final event sample; its value is not stated and directly affects the reported candidate count and efficiency.
axioms (1)
  • domain assumption Inverse beta decay is the dominant and well-understood detection channel for electron antineutrinos in the 8.3-30.8 MeV range inside liquid scintillator.
    Invoked implicitly by the choice of search channel and energy window.

pith-pipeline@v0.9.1-grok · 6116 in / 1534 out tokens · 50650 ms · 2026-06-30T02:13:06.363428+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

45 extracted references · 43 canonical work pages · 2 internal anchors

  1. [1]

    2021, Phys

    Abe, K., Bronner, C., Hayato, Y., et al. 2021, Phys. Rev. D, 104, 122002, doi: 10.1103/PhysRevD.104.122002

    work page doi:10.1103/physrevd.104.122002 2021

  2. [2]

    Search for Diffuse Supernova Neutrino Background with 956.2 days of Super-Kamiokande Gadolinium Dataset

    Abe, K., Abe, S., Asaoka, Y., et al. 2025a, Search for Diffuse Supernova Neutrino Background with 956.2 days of Super-Kamiokande Gadolinium Dataset. https://arxiv.org/abs/2511.02222

    work page internal anchor Pith review Pith/arXiv arXiv

  3. [3]

    2010, Phys

    Abe, S., Enomoto, S., Furuno, K., et al. 2010, Phys. Rev. C, 81, 025807, doi: 10.1103/PhysRevC.81.025807

    work page doi:10.1103/physrevc.81.025807 2010

  4. [4]

    2022, The Astrophysical Journal, 934, 85, doi: 10.3847/1538-4357/ac7a3f

    Abe, S., Asami, S., Eizuka, M., et al. 2022, The Astrophysical Journal, 934, 85, doi: 10.3847/1538-4357/ac7a3f

    work page doi:10.3847/1538-4357/ac7a3f 2022

  5. [5]

    2025b, Phys

    Abe, S., Araki, T., Chiba, K., et al. 2025b, Phys. Rev. Lett., 135, 262501, doi: 10.1103/jkf6-48j8

    work page doi:10.1103/jkf6-48j8

  6. [6]

    Y., Aglietta, M., Antonioli, P., et al

    Agafonova, N. Y., Aglietta, M., Antonioli, P., et al. 2015, The Astrophysical Journal, 802, 47, doi: 10.1088/0004-637X/802/1/47

    work page doi:10.1088/0004-637x/802/1/47 2015

  7. [7]

    2003, Nuclear Instruments and Methods in Physics Research Section A: Accelerators,

    Agostinelli, S., Allison, J., Amako, K., et al. 2003, Nuclear Instruments and Methods in Physics Research Section A: Accelerators,

    2003

  8. [8]

    Spectrometers, Detectors and Associated Equipment, 506, 250, doi: 10.1016/S0168-9002(03)01368-8

    work page doi:10.1016/s0168-9002(03)01368-8

  9. [9]

    2021, Astroparticle Physics, 125, 102509, doi: 10.1016/j.astropartphys.2020.102509

    Agostini, M., Altenm¨ uller, K., Appel, S., et al. 2021, Astroparticle Physics, 125, 102509, doi: 10.1016/j.astropartphys.2020.102509

    work page doi:10.1016/j.astropartphys.2020.102509 2021

  10. [10]

    A., Anderson, C

    Aguilar-Arevalo, A. A., Anderson, C. E., Bazarko, A. O., et al. 2010, Phys. Rev. D, 81, 032001, doi: 10.1103/PhysRevD.81.032001

    work page doi:10.1103/physrevd.81.032001 2010

  11. [11]

    N., Beier, E

    Aharmim, B., Ahmed, S. N., Beier, E. W., et al. 2004, Phys. Rev. D, 70, 093014, doi: 10.1103/PhysRevD.70.093014 18KamLAND collaboration

    work page doi:10.1103/physrevd.70.093014 2004

  12. [12]

    N., Anthony, A

    Aharmim, B., Ahmed, S. N., Anthony, A. E., et al. 2011, The Astrophysical Journal, 728, 83, doi: 10.1088/0004-637X/728/2/83

    work page doi:10.1088/0004-637x/728/2/83 2011

  13. [13]

    A., Aronson, S

    Ahrens, L. A., Aronson, S. H., Connolly, P. L., et al. 1987, Phys. Rev. D, 35, 785, doi: 10.1103/PhysRevD.35.785

    work page doi:10.1103/physrevd.35.785 1987

  14. [14]

    1988, Physics Letters B, 205, 209, doi: 10.1016/0370-2693(88)91651-6

    Volchenko, V. 1988, Physics Letters B, 205, 209, doi: 10.1016/0370-2693(88)91651-6

    work page doi:10.1016/0370-2693(88)91651-6 1988

  15. [15]

    1997, Astroparticle Physics, 7, 357, doi: 10.1016/S0927-6505(97)00035-2

    Kudryavtsev, V., & Sartorelli, G. 1997, Astroparticle Physics, 7, 357, doi: 10.1016/S0927-6505(97)00035-2

    work page doi:10.1016/s0927-6505(97)00035-2 1997

  16. [16]

    2023, The Astrophysical Journal, 953, 151, doi: 10.3847/1538-4357/ace3ba

    Ashida, Y., Nakazato, K., & Tsujimoto, T. 2023, The Astrophysical Journal, 953, 151, doi: 10.3847/1538-4357/ace3ba

    work page doi:10.3847/1538-4357/ace3ba 2023

  17. [17]

    2020, European Physical Journal C, 80, 416, doi: 10.1140/epjc/s10052-020-7977-8

    Askins, M., Bagdasarian, Z., Barros, N., et al. 2020, European Physical Journal C, 80, 416, doi: 10.1140/epjc/s10052-020-7977-8

    work page doi:10.1140/epjc/s10052-020-7977-8 2020

  18. [18]

    S., Ahmad, S., & Singh, S

    Athar, M. S., Ahmad, S., & Singh, S. K. 2007, Phys. Rev. D, 75, 093003, doi: 10.1103/PhysRevD.75.093003

    work page doi:10.1103/physrevd.75.093003 2007

  19. [19]

    2012, Phys

    Bays, K., Iida, T., Abe, K., et al. 2012, Phys. Rev. D, 85, 052007, doi: 10.1103/PhysRevD.85.052007

    work page doi:10.1103/physrevd.85.052007 2012

  20. [20]

    M., Blewitt, G., Bratton, C

    Bionta, R. M., Blewitt, G., Bratton, C. B., et al. 1987, Phys. Rev. Lett., 58, 1494, doi: 10.1103/PhysRevLett.58.1494

    work page doi:10.1103/physrevlett.58.1494 1987

  21. [21]

    B., Casper, D., Ciocio, A., et al

    Bratton, C. B., Casper, D., Ciocio, A., et al. 1988, Phys. Rev. D, 37, 3361, doi: 10.1103/PhysRevD.37.3361

    work page doi:10.1103/physrevd.37.3361 1988

  22. [22]

    Spherical CNNs

    Cohen, T. S., Geiger, M., Koehler, J., & Welling, M. 2018, Spherical CNNs. https://arxiv.org/abs/1801.10130

    work page internal anchor Pith review Pith/arXiv arXiv 2018

  23. [23]

    J., & Cousins, R

    Feldman, G. J., & Cousins, R. D. 1998, Phys. Rev. D, 57, 3873, doi: 10.1103/PhysRevD.57.3873

    work page doi:10.1103/physrevd.57.3873 1998

  24. [24]

    2012, The Astrophysical Journal, 745, 193, doi: 10.1088/0004-637X/745/2/193

    Gando, A., Gando, Y., Ichimura, K., et al. 2012, The Astrophysical Journal, 745, 193, doi: 10.1088/0004-637X/745/2/193

    work page doi:10.1088/0004-637x/745/2/193 2012

  25. [25]

    Reactor On-OffAntineutrino Measurement with KamLAND

    Gando, A., Gando, Y., Hanakago, H., et al. 2013, Phys. Rev. D, 88, 033001, doi: 10.1103/PhysRevD.88.033001

    work page doi:10.1103/physrevd.88.033001 2013

  26. [26]

    2016, Phys

    Gando, A., Gando, Y., Hachiya, T., et al. 2016, Phys. Rev. Lett., 117, 082503, doi: 10.1103/PhysRevLett.117.082503 Geographical Survey Institute of Japan. 1997, Digital Map 50 m Grid (Elevation), unpublished

    work page doi:10.1103/physrevlett.117.082503 2016

  27. [27]

    2023, The Astrophysical Journal Letters, 951, L27, doi: 10.3847/2041-8213/acdc9e

    Harada, M., Abe, K., Bronner, C., et al. 2023, The Astrophysical Journal Letters, 951, L27, doi: 10.3847/2041-8213/acdc9e

    work page doi:10.3847/2041-8213/acdc9e 2023

  28. [28]

    1987, Phys

    Hirata, K., Kajita, T., Koshiba, M., et al. 1987, Phys. Rev. Lett., 58, 1490, doi: 10.1103/PhysRevLett.58.1490

    work page doi:10.1103/physrevlett.58.1490 1987

  29. [29]

    S., Kajita, T., Koshiba, M., et al

    Hirata, K. S., Kajita, T., Koshiba, M., et al. 1988, Phys. Rev. D, 38, 448, doi: 10.1103/PhysRevD.38.448

    work page doi:10.1103/physrevd.38.448 1988

  30. [30]

    2007, Phys

    Honda, M., Kajita, T., Kasahara, K., Midorikawa, S., & Sanuki, T. 2007, Phys. Rev. D, 75, 043006, doi: 10.1103/PhysRevD.75.043006

    work page doi:10.1103/physrevd.75.043006 2007

  31. [31]

    F., & Dwek, E

    Horiuchi, S., Beacom, J. F., & Dwek, E. 2009, Phys. Rev. D, 79, 083013, doi: 10.1103/PhysRevD.79.083013

    work page doi:10.1103/physrevd.79.083013 2009

  32. [32]

    2011, Phys

    Huber, P. 2011, Phys. Rev. C, 84, 024617, doi: 10.1103/PhysRevC.84.024617

    work page doi:10.1103/physrevc.84.024617 2011

  33. [33]

    2007, The Astrophysical Journal, 669, 519, doi: 10.1086/521547 JUNO collaboration

    Ikeda, M., Takeda, A., Fukuda, Y., et al. 2007, The Astrophysical Journal, 669, 519, doi: 10.1086/521547 JUNO collaboration. 2022, Progress in Particle and Nuclear Physics, 123, 103927, doi: 10.1016/j.ppnp.2021.103927

    work page doi:10.1086/521547 2007

  34. [34]

    2003, Phys

    Kamyshkov, Y., & Kolbe, E. 2003, Phys. Rev. D, 67, 076007, doi: 10.1103/PhysRevD.67.076007

    work page doi:10.1103/physrevd.67.076007 2003

  35. [35]

    Kaplinghat, M., Steigman, G., & Walker, T. P. 2000, Phys. Rev. D, 62, 043001, doi: 10.1103/PhysRevD.62.043001

    work page doi:10.1103/physrevd.62.043001 2000

  36. [36]

    2023, Phys

    Li, A., Fu, Z., Grant, C., et al. 2023, Phys. Rev. C, 107, 014323, doi: 10.1103/PhysRevC.107.014323

    work page doi:10.1103/physrevc.107.014323 2023

  37. [37]

    S., Becker-Szendy, R., Bratton, C

    Miller, R. S., Becker-Szendy, R., Bratton, C. B., et al. 1994, The Astrophysical Journal, 428, 629, doi: 10.1086/174272

    work page doi:10.1086/174272 1994

  38. [38]

    2022, The Astrophysical Journal, 938, 35, doi: 10.3847/1538-4357/ac8f41

    Mori, M., Abe, K., Hayato, Y., et al. 2022, The Astrophysical Journal, 938, 35, doi: 10.3847/1538-4357/ac8f41

    work page doi:10.3847/1538-4357/ac8f41 2022

  39. [39]

    Improved Predictions of Reactor Antineutrino Spectra

    Mueller, T. A., Lhuillier, D., Fallot, M., et al. 2011, Phys. Rev. C, 83, 054615, doi: 10.1103/PhysRevC.83.054615

    work page doi:10.1103/physrevc.83.054615 2011

  40. [40]

    2015, The Astrophysical Journal, 804, 75, doi: 10.1088/0004-637X/804/1/75

    Nakazato, K., Mochida, E., Niino, Y., & Suzuki, H. 2015, The Astrophysical Journal, 804, 75, doi: 10.1088/0004-637X/804/1/75

    work page doi:10.1088/0004-637x/804/1/75 2015

  41. [41]

    2013, The Astrophysical Journal Supplement Series, 205, 2, doi: 10.1088/0067-0049/205/1/2

    Nakazato, K., Sumiyoshi, K., Suzuki, H., et al. 2013, The Astrophysical Journal Supplement Series, 205, 2, doi: 10.1088/0067-0049/205/1/2

    work page doi:10.1088/0067-0049/205/1/2 2013

  42. [42]

    2020, Astroparticle Physics, 117, 102404, doi: 10.1016/j.astropartphys.2019.102404

    Novoseltsev, Y., Boliev, M., Dzaparova, I., et al. 2020, Astroparticle Physics, 117, 102404, doi: 10.1016/j.astropartphys.2019.102404

    work page doi:10.1016/j.astropartphys.2019.102404 2020

  43. [43]

    2017, PoS, ICHEP2016, 1161, doi: 10.22323/1.282.1161 AstroNuSearchKam202619

    Ozaki, H., & Shirai, J. 2017, PoS, ICHEP2016, 1161, doi: 10.22323/1.282.1161 AstroNuSearchKam202619

    work page doi:10.22323/1.282.1161 2017

  44. [44]

    2015, in Advances in Neural Information Processing Systems, ed

    Shi, X., Chen, Z., Wang, H., et al. 2015, in Advances in Neural Information Processing Systems, ed. C. Cortes, N. Lawrence, D. Lee, M. Sugiyama, & R. Garnett, Vol. 28 (Curran

    2015

  45. [45]

    1995, Astroparticle Physics, 3, 367, doi: 10.1016/0927-6505(95)00015-9

    Totani, T., & Sato, K. 1995, Astroparticle Physics, 3, 367, doi: 10.1016/0927-6505(95)00015-9

    work page doi:10.1016/0927-6505(95)00015-9 1995