Loading
Life sciences · Preprint
arXiv · September 8, 2026
Early or partial results. Treat as a signal, not a conclusion.
This is a machine-learning feasibility study that applies transfer learning from carbon-12 to six other nuclei to model electron-nucleus cross sections. The text reports that adapted models improve predictions for all targets and shows layer-wise sensitivity varies by nucleus, but does not present quantitative performance metrics, statistical significance tests, or independent validation results.
Computational modelling study; transfer learning with deep neural networks. Electron-nucleus inclusive scattering cross-section measurements across six isotopes. Intervention: Transfer learning domain adaptation: fine-tuning pretrained carbon-12 neural network models on individual target nucleus datasets. Compared with: Phenomenological F1F2 model; carbon-12 baseline performance.
Models fine-tuned on six nuclei (helium-3, lithium-6, oxygen-16, aluminum-27, calcium-40, iron-56) all show improvement over carbon-12 baseline Oxygen requires only shallow adaptation; helium, calcium, and iron require substantially deeper fine-tuning Lithium represents the least robust case because of its limited dataset
Safety was not reported in the material analysed. Check the source before drawing any conclusion about harm.
The source did not state who this applies to in practice.
Early-stage computational study applying transfer learning to nuclear physics modelling with no clinical or patient outcome; demonstrates a methodological approach on multiple nuclei but lacks experimental validation or independent test set performance metrics.
As stated by the source record.
Graded across the dimensions that decide whether you should act, each from what the source actually supports. There is no single score, and where a dimension was not assessed it says so.
What is missing. This record has no reported figures. That is a gap in the analysis, not a judgement about the study.
We apply transfer learning (TL) to construct data-driven models of inclusive electron-nucleus cross sections. Starting from an ensemble of deep neural networks pretrained on \(^{12}\)C data, we fine-tune the models separately for \(^{3}\)He, \(^{6}\)Li, \(^{16}\)O, \(^{27}\)Al, \(^{40}\)Ca, and \(^{56}\)Fe. The resulting models improve for all targets, marginally so for oxygen, where the carbon baseline is already adequate, although their predictive robustness depends on the amount, coverage, and precision of the available target data. We systematically study how model performance depends on the number of fine-tuned layers, on the fraction and selection of the training data, and on the overlap between the source and target kinematic domains. The layer-wise analysis shows that oxygen requires only shallow adaptation, whereas helium, calcium, and iron require substantially deeper fine-tuning. Lithium represents the least robust case because of its limited dataset, while aluminum demonstrates a strong sensitivity to a small subset of highly constraining measurements. For selected kinematic configurations outside the coverage of the carbon training data, the adapted models remain consistent with the measurements within their estimated uncertainties. Finally, we compare the resulting predictions with those of the phenomenological F1F2 model.
Taken from the source record, never inferred. Follow any of these and new work involving them reaches your briefing.