Validation and systematic-uncertainty naming for HS3 models

Determine how to validate that an exported HEP Statistics Serialization Standard (HS3) file faithfully reproduces the original statistical model and how to standardize systematic-uncertainty naming across experiments.

Background

The HEP Statistics Serialization Standard (HS3) is intended to provide a framework-independent JSON description of statistical models, data, and likelihoods. Its adoption requires reliable round-trip conversion between existing RooFit workspaces and HS3 representations, as well as consistent terminology across experiments.

The paper identifies two unresolved issues associated with broader HS3 adoption: establishing validation procedures that verify faithful model reproduction after export, and developing common conventions for naming systematic uncertainties across experiments.

References

Broader adoption also raises open questions, such as how to validate that an exported HS3 file faithfully reproduces the original model, and how to standardize systematic uncertainty naming across experiments.

— Status and Prospects of the HEP Statistical Inference Ecosystem  (2609.19761 - Galli, 17 Sep 2026) in Section 3, “Model preservation and interoperability: HS3”

Broader adoption raises additional open questions, such as how to validate that an exported \HST file faithfully reproduces the original model and how to standardize systematic uncertainty naming across experiments.

— IRIS-HEP 2026 Statistical Ecosystem Blueprint White Paper  (2609.19992 - Feickert et al., 17 Sep 2026) in Section 2, “Statistical Models, Interfaces, Serialization and Publication”