Recent progress in defining jets

From dedicated QCD studies to new physics background estimation, jets will be everywhere at the LHC. In these proceedings, we discuss two important recent series of improvements. In the first one, we introduce new algorithms and new implementations of previously existing algorithms, in order to cure limitations of their predecessors and to satisfy fundamental requirements. In the second part, we show that it is of prime importance to carefully choose the jet definition — algorithm and parameters — to optimise kinematic reconstructions at the LHC. Noticeably, we show that while at scales around 100 GeV, R≃0.5 is an appropriate choice, clustering at the TeV scale requires R≃1 for optimal efficiency. We finally show that our results are valid in the presence of pileup, provided that a subtraction procedure is applied.