Cross-Domain Transfer with Particle Physics Foundation Models: From Jets to Neutrino Interactions

2026-04-14

AI Summary

This study investigates the effective transfer of OmniLearned, a foundation model pretrained on high-energy collider data, to the GeV-scale low-energy neutrino experiment MINERvA, addressing challenges arising from disparities in energy scales, detector technologies, and underlying physical processes. The pretrained model is fine-tuned on neutrino–nucleus scattering events to perform visible energy regression and charged-current pion final-state classification. Results demonstrate that, under identical computational budgets, the transferred model significantly outperforms baselines trained from scratch, providing the first empirical validation of strong generalization and robustness of particle-level foundation models across domains. This advance supports the development of a detector-agnostic inference paradigm in experimental particle physics.

Technology Category

• Machine Learning: Other Foundations of Machine Learning
• Computer Vision: Other Foundations of Computer Vision
• Search and Optimization: Learning to Search

Application Category

• Web Mining and Content Analysis: Large pretrained models with web data
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications

Abstract

Future AI-based studies in particle physics will likely start from a foundation model to accelerate training and enhance sensitivity. As a step towards a general-purpose foundation model for particle physics, we investigate whether the OmniLearned foundation model pre-trained on diverse high-$Q^2$ simulated and real $pp$ and $ep$ collisions can be effectively transferred to a few-GeV fixed-target neutrino experiment. We process MINERvA neutrino--nucleus scattering events and evaluate pre-trained models on two types of tasks: regression of available energy and binary classification of charged-current pion final states ($\mathrm{CC1π^{\pm}}$, $\mathrm{CCNπ^{\pm}}$, and $\mathrm{CC1π^{0}}$). Pre-trained OmniLearned models consistently outperform similarly sized models trained from scratch, achieving better overall performance at the same compute budget, as well as achieving better performance at the same number of training steps. These results suggest that particle-level foundation models acquire inductive biases that generalize across large differences in energy scale, detector technology, and underlying physics processes, pointing toward a paradigm of detector-agnostic inference in particle physics.

Problem

• cross-domain transfer
• foundation models
• neutrino interactions
• particle physics
• domain generalization

Innovation

• foundation model
• cross-domain transfer
• neutrino interactions
• OmniLearned
• detector-agnostic inference