Multiple proton–proton collisions occur at every bunch crossing at the LHC, with the mean
number of interactions reaching about 64 during Run 3 and expected to rise to around 200 at the
High-Luminosity LHC. As a direct consequence, events with multi-jet signatures will appear at
increasingly high rates. To cope with the higher luminosity, efficiently grouping jets according to
their origin along the beamline is crucial, particularly at the trigger level. In this work, a novel
uncertainty-aware jet regression model based on a Deep Sets architecture, DIPz, is introduced
to predict the longitudinal origin of jets along the beamline using the charged-particle tracks
associated with each jet. An event-level discriminant, the Maximum Log Product of Likelihoods
(MLPL), is then constructed by combining the DIPz per-jet predictions. MLPL is optimized to
select events compatible with targeted multi-jet signatures. This combined approach provides a
robust and computationally efficient method for pile-up jet rejection in multi-jet final states, suitable
for real-time event selections at the ATLAS High-Level Trigger.