# ResoSeg: Resonance Tagger using Transformer and Segment Model

arXiv: https://arxiv.org/abs/2609.12610 · HTML: https://mlsanity.com/paper/2609.12610/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 8 of its 35 references that are in the map.
- First posted on arXiv 11 Sep 2026.

## Builds on (most cited first, 8 of 8)

- [Adam: A Method for Stochastic Optimization](https://mlsanity.com/paper/1412.6980/) (2014)
- [PyTorch: An Imperative Style, High-Performance Deep Learning Library](https://mlsanity.com/paper/1912.01703/) (2019)
- [Segment Anything](https://mlsanity.com/paper/2304.02643/) (2023)
- [Multi-Task Learning Using Uncertainty to Weigh Losses for Scene Geometry and Semantics](https://mlsanity.com/paper/1705.07115/) (2017)
- [Gated Attention for Large Language Models: Non-linearity, Sparsity, and Attention-Sink-Free](https://mlsanity.com/paper/2505.06708/) (2025)
- [SPANet: Generalized Permutationless Set Assignment for Particle Physics using Symmetry Preserving Attention](https://mlsanity.com/paper/2106.03898/) (2021)
- [EveNet: A Foundation Model for Particle Collision Data Analysis](https://mlsanity.com/paper/2601.17126/) (2026)
- [First Estimation of Model Parameters for Neutrino-Induced Nucleon Knockout Using Simulation-Based Inference](https://mlsanity.com/paper/2603.09778/) (2026)

## Abstract (arXiv)

Deep learning has been widely applied across many areas of experimental high-energy physics, yet existing models address only event-level classification or object tagging and therefore still require reconstruction algorithms tailored to each decay channel. We present the first application of segmentation to resonance tagging at BESIII and introduce ResoSeg, a deep learning model that jointly performs particle-level segmentation and event-level classification, enabling a one-pass analysis of resonance to anything decays while precisely reconstructing the relevant resonance properties. We demonstrate the reconstruction of ηc with e^+e^-→π^+π^-hc, hc→γηc, ηc→anything. The model is trained on BESIII-ηc dataset, which is constructed with per-track true labels obtained via a Truth-Matching Algorithm. Experimental results show that the average combined efficiency of ResoSeg is more than double that of the conventional 16-channel approach across energy points from 4.19 to 4.60,GeV. The model generalizes to unseen energy points, adapts to other ηc production modes through transfer learning, and remains robust against variations in the ηc mass, width, and branching fractions, providing a general, resonance-aware model applicable beyond ηc and BESIII. The source code is available at https://github.com/oashen/ResoSeg.