# Physics-Informed Graph Neural Networks for Transverse Momentum Estimation in CMS Trigger Systems

arXiv: https://arxiv.org/abs/2507.19205 · HTML: https://mlsanity.com/paper/2507.19205/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 20 of its 53 references that are in the map.
- First posted on arXiv 25 Jul 2025.

## Builds on (most cited first, 20 of 20)

- [Inductive Representation Learning on Large Graphs](https://mlsanity.com/paper/1706.02216/) (2017)
- [Fast Graph Representation Learning with PyTorch Geometric](https://mlsanity.com/paper/1903.02428/) (2019)
- [Dynamic Graph CNN for Learning on Point Clouds](https://mlsanity.com/paper/1801.07829/) (2018)
- [TabNet: Attentive Interpretable Tabular Learning](https://mlsanity.com/paper/1908.07442/) (2019)
- [Fast inference of deep neural networks in FPGAs for particle physics](https://mlsanity.com/paper/1804.06913/) (2018)
- [PyTorch Geometric Temporal: Spatiotemporal Signal Processing with Neural Machine Learning Models](https://mlsanity.com/paper/2104.07788/) (2021)
- [Fast convolutional neural networks on FPGAs with hls4ml](https://mlsanity.com/paper/2101.05108/) (2021)
- [Are Powerful Graph Neural Nets Necessary? A Dissection on Graph Classification](https://mlsanity.com/paper/1905.04579/) (2019)
- [Distance-Weighted Graph Neural Networks on FPGAs for Real-Time Particle Reconstruction in High Energy Physics](https://mlsanity.com/paper/2008.03601/) (2020)
- [TreeLUT: An Efficient Alternative to Deep Neural Networks for Inference Acceleration Using Gradient Boosted Decision Trees](https://mlsanity.com/paper/2501.01511/) (2025)
- [LL-GNN: Low Latency Graph Neural Networks on FPGAs for High Energy Physics](https://mlsanity.com/paper/2209.14065/) (2022)
- [Fast Jet Tagging with MLP-Mixers on FPGAs](https://mlsanity.com/paper/2503.03103/) (2025)
- [A Survey on Graph Classification and Link Prediction based on GNN](https://mlsanity.com/paper/2307.00865/) (2023)
- [End-to-end workflow for machine learning-based qubit readout with QICK and hls4ml](https://mlsanity.com/paper/2501.14663/) (2025)
- [FocusedCleaner: Sanitizing Poisoned Graphs for Robust GNN-based Node Classification](https://mlsanity.com/paper/2210.13815/) (2022)
- [GraphAGILE: An FPGA-based Overlay Accelerator for Low-latency GNN Inference](https://mlsanity.com/paper/2302.01769/) (2023)
- [Hierarchical Graph Neural Networks for Particle Track Reconstruction](https://mlsanity.com/paper/2303.01640/) (2023)
- [Jet tagging algorithm of graph network with HaarPooling message passing](https://mlsanity.com/paper/2210.13869/) (2022)
- [Novel Approaches for ML-Assisted Particle Track Reconstruction and Hit Clustering](https://mlsanity.com/paper/2405.17325/) (2024)
- [Reforge: Low-Latency Distributed GNN Serving with Selective Embedding Recomputation](https://mlsanity.com/paper/2501.08547/) (2025)

## Abstract (arXiv)

Real-time particle transverse momentum (pT) estimation in high-energy physics demands algorithms that are both efficient and accurate under strict hardware constraints. Static machine learning models degrade under high pileup and lack physics-aware optimization, while generic graph neural networks (GNNs) often neglect domain structure critical for robust pT regression. We propose a physics-informed GNN framework that systematically encodes detector geometry and physical observables through four distinct graph construction strategies that systematically encode detector geometry and physical observables: station-as-node, feature-as-node, bending angle-centric, and pseudorapidity (η)-centric representations. This framework integrates these tailored graph structures with a novel Message Passing Layer (MPL), featuring intra-message attention and gated updates, and domain-specific loss functions incorporating pT-distribution priors. Our co-design methodology yields superior accuracy-efficiency trade-offs compared to existing baselines. Extensive experiments on the CMS Trigger Dataset validate the approach: a station-informed EdgeConv model achieves a state-of-the-art MAE of 0.8525 with ≥55% fewer parameters than deep learning baselines, especially TabNet, while an η-centric MPL configuration also demonstrates improved accuracy with comparable efficiency. These results establish the promise of physics-guided GNNs for deployment in resource-constrained trigger systems.