Paper on CMS full event interpretation with MLPF accepted by EPJC

Our paper “Full event interpretation with machine-learning-based particle-flow reconstruction in the CMS detector” has been accepted for publication in the European Physical Journal C (EPJC)!

This work presents the implementation and integration of the machine-learning-based particle-flow (MLPF) reconstruction algorithm within the CMS software framework. We demonstrate learnable full-event reconstruction on GPUs with a 10–20% improvement in jet energy resolution and 5x faster inference compared to standard heuristics.

Read the preprint on arXiv:2601.17554.