Machine-learned particle flow as a foundation model for collider physics
This work provides a concrete demonstration of a foundation model for collider physics, enabling shared representations across reconstruction and analysis tasks, which is a step toward end-to-end pipelines for high-energy physics.
The authors show that a machine-learned particle flow model (MLPF) trained for event reconstruction produces latent representations that can be reused for downstream analysis tasks. Using these representations as additional features improves jet flavor identification, jet energy regression, and missing momentum regression, with a single linear layer outperforming state-of-the-art baselines for missing momentum regression using 35× fewer parameters.
The workflow from particle collision to physics analysis passes through a series of reconstruction steps that are traditionally modular and disconnected, with no shared representation linking low-level detector data to high-level analysis tasks. We show that casting event reconstruction as a machine learning problem naturally produces such a shared representation. We repurpose a machine learning model trained for particle-flow reconstruction (MLPF) to perform three distinct analysis tasks: jet flavor identification, jet energy regression, and missing momentum regression. By appending the per-particle latent representations learned during reconstruction as additional input features, we substantially improve over baselines that use kinematic features alone. We further demonstrate that a single linear layer trained using only the latent representations achieves competitive performance against state-of-the-art baseline architectures, and outperforms the baseline for missing momentum regression with approximately 35 times fewer parameters. These results demonstrate that the latent representations learned during reconstruction encode essential physics information needed for downstream analysis, establishing MLPF as a foundation model and offering a concrete step toward an end-to-end pipeline from detector data to physics analysis.