ATLAS pushes limits on quantum black holes, lepton-flavour violation and Clockwork gravitons
ATLAS at the LHC conducts three new searches for physics beyond the Standard Model, targeting TeV-scale quantum black holes, lepton-flavour-violating resonances, and Clockwork graviton signatures. The analyses use lepton-jet, dilepton-b-jet, dielectron and diphoton final states, employing novel techniques such as wavelet transformations with machine learning to set the strongest limits to date.
ATLAS, the general-purpose detector at the Large Hadron Collider, has released a thesis submitted on 25 Aug 2026 that reports three independent searches for new physics beyond the Standard Model.
Quantum Black Hole Search
The first analysis focuses on hypothetical TeV-scale quantum black holes predicted by extra-dimensional theories that aim to resolve the hierarchy problem. By exploiting a model feature that strongly enhances production rates with increasing centre-of-mass energy, the study uses lepton-plus-jet final states to extend the mass reach. According to the thesis, this approach yields the most stringent limits on the quantum black hole model to date.
Lepton-Flavour-Violating Resonances
The second search addresses anomalies in lepton flavour. It targets heavy resonances that would violate flavour conservation, looking specifically at dilepton plus b-jet final states. This exclusive channel complements earlier inclusive results, providing a direct probe of flavour-violating processes that could explain observed lepton-flavour-universality violations.
Clockwork Graviton Signatures
The third analysis probes Clockwork models, which predict periodic signatures of heavy graviton states in extra-dimensional scenarios. By examining dielectron and diphoton final states, the study introduces a new method that employs wavelet transformations combined with machine learning. This technique allows the identification of periodic structures in the data, leading to the best limits on the Clockwork graviton model reported so far.
Innovative Analysis Techniques
A key feature across all three searches is the use of advanced analysis methods. The wavelet-based approach with machine learning represents a novel strategy for detecting periodic signals in high-energy physics data. In the quantum black hole search, the choice of lepton-plus-jet final states leverages the energy-dependent production enhancement predicted by the model. The lepton-flavour-violating search’s exclusive dilepton-b-jet channel provides a cleaner environment for probing rare flavour-violating decays.
The thesis, available on arXiv, highlights that these combined efforts represent the most comprehensive set of searches for new phenomena using lepton-jet and dilepton-b-jet final states to date. The wavelet-machine-learning method, in particular, opens a new avenue for exploring periodic signatures in collider data, setting a benchmark for future studies of extra-dimensional models.
