{"paper":{"title":"SINAPSE: A lightweight deep learning framework for accurate and explainable neutron-$\\gamma$ discrimination","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"A lightweight dual-branch neural network denoises low-charge waveforms and classifies neutrons versus gammas with calibrated probabilities.","cross_cats":["physics.data-an"],"primary_cat":"physics.ins-det","authors_text":"Adrien Matta, Audrey Chatillon, Beno\\^it Mauss, Charl\\`ene Surault, Cyril Lenain, David Etasse, David Regnier, Jason Surbrook, Julien Taieb, Matthew Devlin, Owen Syrett, Patrick Copp, Pierre Morfouace, Thomas Carreau","submitted_at":"2026-05-13T14:53:47Z","abstract_excerpt":"Traditionally, neutron-$\\gamma$ discrimination in organic scintillators relies on techniques such as time-of-flight (ToF) selection and pulse-shape discrimination (PSD). However, particle identification through graphical cuts remains challenging in the low-charge regime due to poor signal-to-noise ratios (SNR). In this work, we propose SINAPSE, a lightweight deep learning framework for accurate and explainable neutron-$\\gamma$ discrimination in the low-charge regime. The framework employs a dual-branch architecture that combines a 1-dimensional convolutional autoencoder for waveform denoising "},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"SINAPSE achieves superior denoising performance compared to conventional digital signal processing techniques, and outputs well-calibrated probabilities, consistent with traditional graphical cuts.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"That random augmentations applied to high-SNR waveforms faithfully reproduce the noise statistics and pulse-shape distortions present in real low-charge experimental data.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"SINAPSE uses a dual-branch neural network with a 1D convolutional autoencoder for denoising and a classifier for neutron-gamma discrimination, trained via random augmentations on high-SNR data and validated with SHAP explanations.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"A lightweight dual-branch neural network denoises low-charge waveforms and classifies neutrons versus gammas with calibrated probabilities.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"084fe8c56749b22611a1a54b8e0a7f21499ad7ac8e0bd1f6a5c5a9b1251db1c3"},"source":{"id":"2605.13627","kind":"arxiv","version":1},"verdict":{"id":"3fe24b94-bd77-4efd-b404-1984840a2d6b","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-14T17:47:40.108538Z","strongest_claim":"SINAPSE achieves superior denoising performance compared to conventional digital signal processing techniques, and outputs well-calibrated probabilities, consistent with traditional graphical cuts.","one_line_summary":"SINAPSE uses a dual-branch neural network with a 1D convolutional autoencoder for denoising and a classifier for neutron-gamma discrimination, trained via random augmentations on high-SNR data and validated with SHAP explanations.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"That random augmentations applied to high-SNR waveforms faithfully reproduce the noise statistics and pulse-shape distortions present in real low-charge experimental data.","pith_extraction_headline":"A lightweight dual-branch neural network denoises low-charge waveforms and classifies neutrons versus gammas with calibrated probabilities."},"references":{"count":34,"sample":[{"doi":"","year":null,"title":"R. C. Haight, H. Y. Lee, T. N. Taddeucci, J. M. O’Donnell, B. A. Perdue, N. Fotiades, M. Devlin, J. L. Ullmann, A. Laptev, T. Bredeweg,et al., Two detector arrays for fast neutrons at lansce, Journal ","work_id":"00c9259b-ffe7-4b3b-a126-2eb552bded29","ref_index":1,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2014,"title":"T. Mart´ ınez, D. Cano-Ott, J. Castilla, A. Garcia, J. Marin, G. Martinez, E. Mendoza, C. Santos, F. Tera, D. Villamarin,et al., Monster: a tof spectrometer forβ- delayed neutron spectroscopy, Nuclear","work_id":"1bc42e63-a69c-476f-8211-8cad910f9136","ref_index":2,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2023,"title":"P. Talou and R. Vogt,Nuclear fission: theories, experi- ments and applications(Springer Nature, 2023)","work_id":"b9ccb4e4-b4f8-4c78-99d9-03c6a71d8575","ref_index":3,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":1979,"title":"Brooks, Development of organic scintillators, Nuclear Instruments and Methods162, 477 (1979)","work_id":"9de5ede2-7b29-4176-a521-addfc6ed40e8","ref_index":4,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2021,"title":"X. Fabian, G. Baulieu, L. Ducroux, O. St´ ezowski, A. Boujrad, E. Cl´ ement, S. Coudert, G. de France, N. Er- duran, S. Ert¨ urk,et al., Artificial neural networks for neutron/γdiscrimination in the n","work_id":"7accc238-bca1-4b4c-921c-5e35fac0ddd4","ref_index":5,"cited_arxiv_id":"","is_internal_anchor":false}],"resolved_work":34,"snapshot_sha256":"3d0e35028af51791f54338478b3ccef8b43afa025484bbe57b5d1ad3743bfbae","internal_anchors":1},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"}