Transformer-based Reconstruction for Electromagnetic Calorimeters in Future LHC Upgrade Experiments

Description

Electromagnetic calorimeter reconstruction is a critical component of precision measurements involving neutral particles such as photons and neutral pions (π⁰). The achievable energy resolution directly impacts the sensitivity of physics analyses relying on these final states, including rare decays and CP violation measurements.

In the context of future LHC upgrades, calorimeter reconstruction must satisfy increasingly stringent real-time constraints, making both reconstruction quality and inference performance essential. Transformer-based machine learning models have recently emerged as a promising technology for modeling complex detector responses and long-range correlations, with potential advantages in reconstruction accuracy and scalability.

The goal of this project is to design, implement, and benchmark a Transformer-based reconstruction pipeline for electromagnetic calorimeters, focusing on energy resolution and inference performance. The developed approach will be quantitatively compared to existing standard reconstruction algorithms and GNN-based methods. The project emphasizes software implementation, validation, and benchmarking, rather than open-ended machine learning research, making it well suited for the GSoC timeline.

Task Ideas

Expected Results and Milestones

Core deliverables

Stretch goals (depending on progress)

Requirements

AI Policy

AI assistance is allowed for this contribution. The applicant takes full responsibility for all code and results, disclosing AI use for non-routine tasks (algorithm design, architecture, complex problem-solving). Routine tasks (grammar, formatting, style) do not require disclosure.

How to Apply

Email mentors with a brief background and interest in ML/particle physics. Please include “gsoc26” in the subject line. Mentors will provide an evaluation task after submission.

Resources

Mentors

Additional Information

Corresponding Project

Participating Organizations