The new nonclosure loss term, made differentiable with a sigmoid approximation, lets the neural network optimize the ABCD background estimate directly, improving closure and training stability.
A Stealth Supersymmetry Sampler
1 Pith paper cite this work. Polarity classification is still indexing.
abstract
The LHC has strongly constrained models of supersymmetry with traditional missing energy signatures. We present a variety of models that realize the concept of Stealth Supersymmetry, i.e. models with R-parity in which one or more nearly-supersymmetric particles (a "stealth sector") lead to collider signatures with only a small amount of missing energy. The simplest realization involves low-scale supersymmetry breaking, with an R-odd particle decaying to its superpartner and a soft gravitino. We clarify the stealth mechanism and its differences from compressed supersymmetry and explain the requirements for stealth models with high-scale supersymmetry breaking, in which the soft invisible particle is not a gravitino. We also discuss new and distinctive classes of stealth models that couple through a baryon portal or Z' gauge interactions. Finally, we present updated limits on stealth supersymmetry in light of current LHC searches.
citation-role summary
citation-polarity summary
fields
hep-ex 1years
2025 1verdicts
CONDITIONAL 1roles
background 1polarities
background 1representative citing papers
citing papers explorer
-
Machine learning method for enforcing variable independence in background estimation with LHC data: ABCDisCoTEC
The new nonclosure loss term, made differentiable with a sigmoid approximation, lets the neural network optimize the ABCD background estimate directly, improving closure and training stability.