The CMS experiment at CERN has conducted a search for light-weight hypothetical particles, specifically axion-like particles (ALPs), that may result from Higgs boson decays. These ALPs are theorised to decay into electron-positron pairs, which present a significant challenge for detection due to their merged experimental footprints. To overcome this, the CMS Collaboration developed a new multivariate identification technique using machine learning and precise tracking data. While no significant excess beyond the Standard Model was observed, the research established new stringent limits on exotic Higgs boson decays, thus providing crucial constraints on theoretical models of ALPs and demonstrating the LHC's unique capacity for exploring new physics.

The ATLAS Experiment at CERN is actively searching for evidence of a "dark sector," a theoretical realm of particles and forces that interact minimally with known matter, potentially explaining dark matter. Researchers are specifically investigating "dark jets," which are sprays of dark-sector particles that could partially decay into visible particles or vanish during high-energy collisions. To distinguish these elusive signals from background noise, the experiment employs advanced artificial intelligence and machine learning models to analyse vast amounts of data. Two primary search strategies are highlighted: one targeting "semi-visible jets," which exhibit missing momentum due to undetected dark particles, and another focusing on "emerging jets," where particles appear to materialise some distance from the collision point. These studies utilise sophisticated analysis techniques, including neural networks and unique trigger systems, to place exclusion limits on theoretical mediator particles that could link the Standard Model to the dark sector.

More info here: https://arxiv.org/abs/2505.02429

In this episode, Bob and Alice discuss this CMS result: https://cms-results.web.cern.ch/cms-results/public-results/preliminary-results/JME-24-001/index.html