Self-Supervised Learning for Robust Resonance Mass Regression in Cascade Decays
Abstract
Reconstructing the mass of a heavy resonance from its decay products with missing energy is one of the central tasks that directly determine the sensitivity in new physics searches at collider experiments. Supervised learning approaches to this problem often struggle to generalize well due to the presence of various systematic uncertainties and distribution shifts. Exhausting all possible variations in the labeled data can be very compute-intensive, while a failure of the model to generalize can corrupt the reconstructed resonance widths that are critical in peak-hunting analyses. In this work, following the foundation model paradigm, we use a self-supervised approach to pre-train a transformer encoder with VICReg to learn an embedding invariant to various corruptions, then fine-tune it for mass regression on a heavy resonance with masses ranging from 2.5 to 6.5 TeV and a SUSY-like cascade decay into an eleven-body final state. We show that the pre-trained model reconstructs sharper resonance peaks and has a more stable performance under various realistic corruptions, compared to a supervised model of the same architecture trained on the same augmented data from scratch.
Related
- Deep learning-based self-supervised detection and localization of magnetic resonance imaging anomalies in artificial intelligence-supported radiotherapy workflows
- Comment on "Supervised Learning with Quantum Hamiltonian Embeddings" (arXiv:2610.26833)
- Improving Signals, Declining Encoder: A Known-Answer Test of Label-Free Monitoring in Self-Supervised Learning
- Lyman Break Galaxy selection and redshift measurement with supervised contrastive learning
- Exploring Joint-Embedding Predictive Architectures for Self-Supervised Representation Learning of Powder X-ray Diffraction Patterns
BibTeX
@misc{tsoi2026selfsupervised,
title = {{Self-Supervised Learning for Robust Resonance Mass Regression in Cascade Decays}},
author = {Ho Fung Tsoi and Alex Yang and Luis Felipe Gutierrez Zagazeta and Shion Chen and Dylan Rankin},
howpublished = {arXiv},
year = {2026},
month = oct,
doi = {10.48550/arxiv.2609.17726},
eprint = {2609.17726},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2609.17726},
note = {CC BY 4.0},
}
RIS
TY - UNPB
TI - Self-Supervised Learning for Robust Resonance Mass Regression in Cascade Decays
AU - Ho Fung Tsoi
AU - Alex Yang
AU - Luis Felipe Gutierrez Zagazeta
AU - Shion Chen
AU - Dylan Rankin
PY - 2026
DA - 2026/10/26
PB - arXiv
DO - 10.48550/arxiv.2609.17726
UR - https://arxiv.org/abs/2609.17726
AB - Reconstructing the mass of a heavy resonance from its decay products with missing energy is one of the central tasks that directly determine the sensitivity in new physics searches at collider experiments. Supervised learning approaches to this problem often struggle to generalize well due to the presence of various systematic uncertainties and distribution shifts. Exhausting all possible variations in the labeled data can be very compute-intensive, while a failure of the model to generalize can corrupt the reconstructed resonance widths that are critical in peak-hunting analyses. In this work, following the foundation model paradigm, we use a self-supervised approach to pre-train a transformer encoder with VICReg to learn an embedding invariant to various corruptions, then fine-tune it for mass regression on a heavy resonance with masses ranging from 2.5 to 6.5 TeV and a SUSY-like cascade decay into an eleven-body final state. We show that the pre-trained model reconstructs sharper resonance peaks and has a more stable performance under various realistic corruptions, compared to a supervised model of the same architecture trained on the same augmented data from scratch.
ER -