Probing the fundamental limit of jet tagging with surrogate reference models
Physical Review D · 16.11.2026 · 114 (9)
DOI: 10.1103/rcpy-nr72 ↗Kurz gefasst
Fachartikel in der Zeitschrift Physical Review D, online erschienen am 16.11.2026. Verfasst von Ian Pang und 4 weiteren Autor:innen. Die Arbeit nennt 35 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: U.S. Department of Energy, Deutsche Forschungsgemeinschaft, National Energy Research Scientific Computing Center.
Abstract (Originalsprache)
Beyond the practical goal of improving search and measurement sensitivity through better jet tagging algorithms, there is a deeper question: what are their upper performance limits? Generative surrogate models with learned likelihood functions offer a new approach to this problem, provided the surrogate correctly captures the underlying data distribution. In this work, we introduce the SUrrogate ReFerence (SURF) method, a new approach to validating generative models. This framework enables exact Neyman-Pearson tests by training the target model on samples from another tractable surrogate, which is itself trained on real data. We argue that the EPiC-FM generative model is a valid surrogate reference for jets and apply SURF to show that modern jet taggers may already be operating close to the true statistical limit. By contrast, we find that autoregressive generative pretrained transformer models unphysically exaggerate top vs QCD separation power encoded in the surrogate reference, implying that they are giving a misleading picture of the fundamental limit.
Förderung
- U.S. Department of Energy
- Deutsche Forschungsgemeinschaft
- National Energy Research Scientific Computing Center
- Office of Science
Zitieren
APA
Pang, I., Faroughy, D. A., Shih, D., Das, R., & Kasieczka, G. (2026). Probing the fundamental limit of jet tagging with surrogate reference models. Physical Review D, 114(9). https://doi.org/10.1103/rcpy-nr72
Vancouver
Pang I, Faroughy DA, Shih D, Das R, Kasieczka G. Probing the fundamental limit of jet tagging with surrogate reference models. Physical Review D. 2026;114(9). doi:10.1103/rcpy-nr72
BibTeX
@article{Pang2026,
author = {Pang, Ian and Faroughy, Darius A. and Shih, David and Das, Ranit and Kasieczka, Gregor},
title = {Probing the fundamental limit of jet tagging with surrogate reference models},
journal = {Physical Review D},
year = {2026},
volume = {114},
number = {9},
doi = {10.1103/rcpy-nr72}
}
EUVIMED