Speaker
Description
In many BSM scenarios, dark matter candidates manifest as long-lived particles (LLPs), arising from suppressed couplings or small mass splittings. These scenarios are probed through LLP searches at colliders and fixed-target experiments, complemented by dark matter constraints from direct, indirect, and cosmological observations. However, results across these programmes are reported in heterogeneous parameterisations, such as cross-sections, event rates, lifetimes, branching fractions, or model-dependent couplings (e.g. kinetic mixing or portal interactions), complicating reinterpretation and consistent global constraints.
DarkMatterLive is a collaborative framework for building and sharing interactive visualisations of dark matter constraints, including LLP and dark sector models, based on published experimental and observational results. It brings together limits that are otherwise difficult to compare, reinterpreting and renormalising them in a consistent, model-aware parameter space. Where possible, constraints are recast in standard coupling-mass-lifetime variables following widely used conventions, such as the Beyond Collider benchmarks, while preserving clear links to the original sources, assumptions, and caveats.
The platform already supports LLP use-cases such as dark photon , where constraints on kinetic mixing are mapped into lifetime-mass space relevant for displaced signatures. Hosted on CERN PaaS cloud infrastructure, it is publicly available through the https://darkmatter.web.cern.ch/ web interface, with a citable release archived as 10.5281/zenodo.18135859. It also connects to community resources such as the HEPData portal, helping organise new inputs within a transparent review workflow.
To make contributions easier, DarkMatterLive provides a data-ingestion workflow based on CSV curve data and JSON metadata. It supports contributions from multiple communities, where new results and key reinterpretation and renormalisation factors are documented and peer-reviewed through GitHub pull requests or CERN GitLab merge requests before inclusion. Code and datasets are openly maintained, supporting transparency and shared ownership. Additional AI-MCP-based ingestion modules are being integrated into platform workflows.
| Submitted on behalf of ... | Submitted centrally on behalf of the Collaboration |
|---|