PT Journal AU ATLAS Collaboration (Aad, Gea Akiot, A Amos, KR Aparisi Pozo, JA Bailey, AJ Bouchhar, N Cabrera Urban, S Cantero, J Cardillo, F Castillo Gimenez, V Chitishvili, M Costa, MJ Didenko Escobar, C Fiorini, L Fullana Torregrosa, E Fuster, J Garcia, C Garcia Navarro, JE Gomez Delegido, AJ Gonzalez de la Hoz, S Gonzalvo Rodriguez, GR Guerrero Rojas, JGR Lacasta, C Marti-Garcia, S Martinez Agullo, P Miralles Lopez, M Mitsou, VA Monsonis Romero, L Moreno Llacer, M Munoz Perez, D Navarro-Gonzalez, J Poveda, J Prades IbaƱez, A Rubio Jimenez, A Ruiz-Martinez, A Sabatini, P Salt, J Sanchez Sebastian, V Sayago Galvan, I Senthilkumar, V Soldevila, U Sanchez, J Torro Pastor, E Valero, A Valiente Moreno, E Valls Ferrer, JA Varriale, L Villaplana Perez, M Vos, M TI Fast b-tagging at the high-level trigger of the ATLAS experiment in LHC Run 3 SO Journal of Instrumentation JI J. Instrum. PY 2023 BP P11006 - 38pp VL 18 IS 11 DI 10.1088/1748-0221/18/11/P11006 LA English DE Trigger algorithms; Trigger concepts and systems (hardware and software) AB The ATLAS experiment relies on real-time hadronic jet reconstruction and b-tagging to record fully hadronic events containing b-jets. These algorithms require track reconstruction, which is computationally expensive and could overwhelm the high-level-trigger farm, even at the reduced event rate that passes the ATLAS first stage hardware-based trigger. In LHC Run 3, ATLAS has mitigated these computational demands by introducing a fast neural-network-based b-tagger, which acts as a low-precision filter using input from hadronic jets and tracks. It runs after a hardware trigger and before the remaining high-level-trigger reconstruction. This design relies on the negligible cost of neural-network inference as compared to track reconstruction, and the cost reduction from limiting tracking to specific regions of the detector. In the case of Standard Model HH -> b (b) over barb (b) over bar, a key signature relying on b-jet triggers, the filter lowers the input rate to the remaining high-level trigger by a factor of five at the small cost of reducing the overall signal efficiency by roughly 2%. ER