Searching for beyond the standard model (BSM) physics is one of the primary tasks of the LHC, as many of the mysteries, such as the origin of dark matter, can not be explained by the standard model (SM). Both the CMS and ATLAS experiments have explored an impressive amount of hypotheses and models comprehensively. Many exotic signatures, such as long-lived particles, ultra energetic heavy flavour jets and jets with anomalous structures have been explored. This type of searches is often limited by the detector performance as originally the algorithms were designed to reconstruct and identify signatures expected in the SM. Recent developments in ATLAS, such as the new large radius tracking (LRT) algorithm and the new graph neural network (GNN) based flavour tagging algorithms, make it possible to allow the standard ATLAS algorithms to probe those exotic signatures. In this talk, I will have a brief introduction to the above algorithms, and discuss its impact on relevant BSM searches. The exotic signature will become ordinary in the foreseen future, and we are ready to take new challenges.
