# Explaining machine-learned particle-flow reconstruction

arXiv: https://arxiv.org/abs/2111.12840 · HTML: https://mlsanity.com/paper/2111.12840/

## Key facts

- Cited by 12 papers in the map (data through 14 Oct 2026).
- Builds on 12 of its 61 references that are in the map.
- First posted on arXiv 24 Nov 2021.

## Builds on (most cited first, 12 of 12)

- [Fast Graph Representation Learning with PyTorch Geometric](https://mlsanity.com/paper/1903.02428/) (2019)
- [Convolutional Neural Networks on Graphs with Fast Localized Spectral Filtering](https://mlsanity.com/paper/1606.09375/) (2016)
- [GNNExplainer: Generating Explanations for Graph Neural Networks](https://mlsanity.com/paper/1903.03894/) (2019)
- [Higher-Order Explanations of Graph Neural Networks via Relevant Walks](https://mlsanity.com/paper/2006.03589/) (2020)
- [ParticleNet: Jet Tagging via Particle Clouds](https://mlsanity.com/paper/1902.08570/) (2019)
- [Particle Cloud Generation with Message Passing Generative Adversarial Networks](https://mlsanity.com/paper/2106.11535/) (2021)
- [Learning representations of irregular particle-detector geometry with distance-weighted graph networks](https://mlsanity.com/paper/1902.07987/) (2019)
- [MLPF: Efficient machine-learned particle-flow reconstruction using graph neural networks](https://mlsanity.com/paper/2101.08578/) (2021)
- [Object condensation: one-stage grid-free multi-object reconstruction in physics detectors, graph and image data](https://mlsanity.com/paper/2002.03605/) (2020)
- [Charged particle tracking via edge-classifying interaction networks](https://mlsanity.com/paper/2103.16701/) (2021)
- [Explainable AI for ML jet taggers using expert variables and layerwise relevance propagation](https://mlsanity.com/paper/2011.13466/) (2020)
- [Track Seeding and Labelling with Embedded-space Graph Neural Networks](https://mlsanity.com/paper/2007.00149/) (2020)

## Built on by (most cited first, 12 of 12)

- [Interpretable Machine Learning in Physics: A Review](https://mlsanity.com/paper/2503.23616/) (2025)
- [Graph Neural Networks in Particle Physics: Implementations, Innovations, and Challenges](https://mlsanity.com/paper/2203.12852/) (2022)
- [Explainable AI for High Energy Physics](https://mlsanity.com/paper/2206.06632/) (2022)
- [A Detailed Study of Interpretability of Deep Neural Network based Top Taggers](https://mlsanity.com/paper/2210.04371/) (2022)
- [Progress towards an improved particle flow algorithm at CMS with machine learning](https://mlsanity.com/paper/2303.17657/) (2023)
- [Do graph neural networks learn traditional jet substructure?](https://mlsanity.com/paper/2211.09912/) (2022)
- [FAIR AI Models in High Energy Physics](https://mlsanity.com/paper/2212.05081/) (2022)
- [Mixture-of-Experts Graph Transformers for Interpretable Particle Collision Detection](https://mlsanity.com/paper/2501.03432/) (2025)
- [Data Science and Machine Learning in Education](https://mlsanity.com/paper/2207.09060/) (2022)
- [Learning Tree Structures from Leaves For Particle Decay Reconstruction](https://mlsanity.com/paper/2208.14924/) (2022)
- [Graph Structure from Point Clouds: Geometric Attention is All You Need](https://mlsanity.com/paper/2307.16662/) (2023)
- [Dissecting Jet-Tagger Through Mechanistic Interpretability](https://mlsanity.com/paper/2605.09881/) (2026)

## Abstract (arXiv)

The particle-flow (PF) algorithm is used in general-purpose particle detectors to reconstruct a comprehensive particle-level view of the collision by combining information from different subdetectors. A graph neural network (GNN) model, known as the machine-learned particle-flow (MLPF) algorithm, has been developed to substitute the rule-based PF algorithm. However, understanding the model's decision making is not straightforward, especially given the complexity of the set-to-set prediction task, dynamic graph building, and message-passing steps. In this paper, we adapt the layerwise-relevance propagation technique for GNNs and apply it to the MLPF algorithm to gauge the relevant nodes and features for its predictions. Through this process, we gain insight into the model's decision-making.