In recent years, LHC experiments are increasingly publishing the full likelihood models for statistical analysis in HEPData. For the purporse of reproducing and reinterpreting such analyses, it is essential to understand the contents of these statistical workspaces. As part of the validation process for an ATLAS combination of searches for Beyond-Standard-Model particles, I have developed an...
ATLAS and CMS have conducted two pairs of broadly equivalent analyses, and find excesses in all four cases. This consists of searches for soft leptons (originally in supersymmetric models with compressed spectra) and monojets. I will describe recent work on trying to interpret these excesses, and quantify their significance, in terms of both supersymmetric and non-supersymmetric models, and to...
The landscape of frameworks for (re)implementing HEP analyses for reinterpretation is, fortunately, diverse. One advantage of this diversity is that multiple reimplementations of the same HEP analysis in different frameworks can be cross-validated. On the downside, statistical combinations of analyses across different frameworks must carefully
avoid double-counting events, which could...
Following on from discussions at previous reinterpretation forums on the reinterpretation of ML-based searches; we look at how one of these (ATLAS-SUSY-2018-30) has been used as a test-bed for studies of ML-based reweighting, which can reduce the computational load of studying large parameter spaces. Some new features of YODA 2 which allowed the project to run in Rivet will also be highlighted.
An update on Contur (Contraints On New Theories Using Rivet) developments
