Evidence for tt̄tt̄ production in the multilepton final state in proton-proton collisions at $\sqrt{s}=13$ TeV with the ATLAS detector

ATLAS-CONF-2020-013

27 May 2020

These preliminary results are superseded by the following paper:

TOPQ-2018-05
ATLAS recommends to use the results from the paper.

Content

Preview

Main document (CDS record), Physics Briefing - internal

pdf from CDS

Figures Tables

Abstract

A search for four-top-quark production using proton-proton collision data at a centre-of-mass energy of 13 TeV collected by the ATLAS detector corresponding to an integrated luminosity of 139 fb$^{-1}$ is presented. Events are selected if they contain a same-sign lepton pair or at least three leptons (electrons or muons). Jet multiplicity, jet flavour and event kinematics are used to separate signal from the background through a multivariate discriminant, and dedicated control regions are used to constrain the dominant backgrounds. The four-top-quark production cross section is measured to be $24^{+7}_{-6}$ fb, corresponding to an observed (expected) signal significance of 4.3 (2.4) standard deviations and providing evidence for this process.

Figures

Figure 01a:
Examples of Feynman diagrams for SM tt̄tt̄ production at leading order in QCD.

png (137kB) pdf (26kB)

Figure 01b:
Examples of Feynman diagrams for SM tt̄tt̄ production at leading order in QCD.

png (144kB) pdf (26kB)

Figure 02a:
Pre-fit comparison between data and prediction in the signal region for some of the input variables used to train the multivariate discriminant: the pseudo-continuous b-tagging discriminant score summed over all the jets in the event (a) and the minimum distance between two leptons among all possible pairs (b). The band includes the total uncertainty on the pre-fit computation. The ratio of the data to the total pre-fit expectation is shown in the lower panel. The first and last bin contain underflow and overflow events, respectively.

png (162kB) pdf (22kB)

Figure 02b:
Pre-fit comparison between data and prediction in the signal region for some of the input variables used to train the multivariate discriminant: the pseudo-continuous b-tagging discriminant score summed over all the jets in the event (a) and the minimum distance between two leptons among all possible pairs (b). The band includes the total uncertainty on the pre-fit computation. The ratio of the data to the total pre-fit expectation is shown in the lower panel. The first and last bin contain underflow and overflow events, respectively.

png (158kB) pdf (24kB)

Figure 03:
Comparison between data and prediction after the fit ("Post-Fit") for the distribution of the BDT score in the SR. The band includes the total uncertainty on the post-fit computation. The ratio of the data to the total post-fit computation is shown in the lower panel.

png (147kB) pdf (25kB)

Figure 04a:
Comparison between data and prediction after the fit ("Post-Fit") for the distributions of the variables used in the fit in each CR. The band includes the total uncertainty on the post-fit computation. The ratio of the data to the total post-fit computation is shown in the lower panel. The first and last bin contain underflow and overflow events, respectively.

png (117kB) pdf (17kB)

Figure 04b:
Comparison between data and prediction after the fit ("Post-Fit") for the distributions of the variables used in the fit in each CR. The band includes the total uncertainty on the post-fit computation. The ratio of the data to the total post-fit computation is shown in the lower panel. The first and last bin contain underflow and overflow events, respectively.

png (107kB) pdf (17kB)

Figure 04c:
Comparison between data and prediction after the fit ("Post-Fit") for the distributions of the variables used in the fit in each CR. The band includes the total uncertainty on the post-fit computation. The ratio of the data to the total post-fit computation is shown in the lower panel. The first and last bin contain underflow and overflow events, respectively.

png (146kB) pdf (20kB)

Figure 04d:
Comparison between data and prediction after the fit ("Post-Fit") for the distributions of the variables used in the fit in each CR. The band includes the total uncertainty on the post-fit computation. The ratio of the data to the total post-fit computation is shown in the lower panel. The first and last bin contain underflow and overflow events, respectively.

png (141kB) pdf (20kB)

Figure 05a:
Post-fit comparison between data and prediction in the ttW validation region for the multiplicity of jets (a) and the BDT score (b). The y-axis label N₊ - N_- represents the difference between the number of events with a positive sum and the number of events with a negative sum of the charges of the selected leptons. of the selected leptons and the number of events with a negative sum. The band includes the total uncertainty on the post-fit computation. The ratio of the data to the total post-fit computation is shown in the lower panel. The first and last bin contain underflow and overflow events, respectively.

png (132kB) pdf (18kB)

Figure 05b:
Post-fit comparison between data and prediction in the ttW validation region for the multiplicity of jets (a) and the BDT score (b). The y-axis label N₊ - N_- represents the difference between the number of events with a positive sum and the number of events with a negative sum of the charges of the selected leptons. of the selected leptons and the number of events with a negative sum. The band includes the total uncertainty on the post-fit computation. The ratio of the data to the total post-fit computation is shown in the lower panel. The first and last bin contain underflow and overflow events, respectively.

png (132kB) pdf (19kB)

Figure 06a:
Post-fit comparison between data and prediction for signal region events for the distributions of: the sum of b-tagging pseudo-continuous scores of the jets in the event (a), the minimum distance between two leptons among all possible pairs (b), the multiplicity of jets (c) and the the multiplicity of b-tag jets (d). The band includes the total uncertainty on the post-fit computation. The ratio of the data to the total post-fit computation is shown in the lower panel. The first and last bin contain underflow and overflow events, respectively.

png (147kB) pdf (22kB)

Figure 06b:
Post-fit comparison between data and prediction for signal region events for the distributions of: the sum of b-tagging pseudo-continuous scores of the jets in the event (a), the minimum distance between two leptons among all possible pairs (b), the multiplicity of jets (c) and the the multiplicity of b-tag jets (d). The band includes the total uncertainty on the post-fit computation. The ratio of the data to the total post-fit computation is shown in the lower panel. The first and last bin contain underflow and overflow events, respectively.

png (138kB) pdf (23kB)

Figure 06c:
Post-fit comparison between data and prediction for signal region events for the distributions of: the sum of b-tagging pseudo-continuous scores of the jets in the event (a), the minimum distance between two leptons among all possible pairs (b), the multiplicity of jets (c) and the the multiplicity of b-tag jets (d). The band includes the total uncertainty on the post-fit computation. The ratio of the data to the total post-fit computation is shown in the lower panel. The first and last bin contain underflow and overflow events, respectively.

png (133kB) pdf (19kB)

Figure 06d:
Post-fit comparison between data and prediction for signal region events for the distributions of: the sum of b-tagging pseudo-continuous scores of the jets in the event (a), the minimum distance between two leptons among all possible pairs (b), the multiplicity of jets (c) and the the multiplicity of b-tag jets (d). The band includes the total uncertainty on the post-fit computation. The ratio of the data to the total post-fit computation is shown in the lower panel. The first and last bin contain underflow and overflow events, respectively.

png (130kB) pdf (18kB)

Figure 07a:
Post-fit comparison between data and prediction for signal region events with a BDT score greater than zero for the distributions of: the sum of b-tagging pseudo-continuous scores of the jets in the event (a), the minimum distance between two leptons among all possible pairs (b), the multiplicity of jets (c) and the the multiplicity of b-tag jets (d). The band includes the total uncertainty on the post-fit computation. The ratio of the data to the total post-fit computation is shown in the lower panel. The first and last bin contain underflow and overflow events, respectively.

png (136kB) pdf (21kB)

Figure 07b:
Post-fit comparison between data and prediction for signal region events with a BDT score greater than zero for the distributions of: the sum of b-tagging pseudo-continuous scores of the jets in the event (a), the minimum distance between two leptons among all possible pairs (b), the multiplicity of jets (c) and the the multiplicity of b-tag jets (d). The band includes the total uncertainty on the post-fit computation. The ratio of the data to the total post-fit computation is shown in the lower panel. The first and last bin contain underflow and overflow events, respectively.

png (135kB) pdf (23kB)

Figure 07c:
Post-fit comparison between data and prediction for signal region events with a BDT score greater than zero for the distributions of: the sum of b-tagging pseudo-continuous scores of the jets in the event (a), the minimum distance between two leptons among all possible pairs (b), the multiplicity of jets (c) and the the multiplicity of b-tag jets (d). The band includes the total uncertainty on the post-fit computation. The ratio of the data to the total post-fit computation is shown in the lower panel. The first and last bin contain underflow and overflow events, respectively.

png (131kB) pdf (19kB)

Figure 07d:
Post-fit comparison between data and prediction for signal region events with a BDT score greater than zero for the distributions of: the sum of b-tagging pseudo-continuous scores of the jets in the event (a), the minimum distance between two leptons among all possible pairs (b), the multiplicity of jets (c) and the the multiplicity of b-tag jets (d). The band includes the total uncertainty on the post-fit computation. The ratio of the data to the total post-fit computation is shown in the lower panel. The first and last bin contain underflow and overflow events, respectively.

png (131kB) pdf (18kB)

Figure FourTopsEvent:
Four-top candidate from 2018. Event display of a candidate four-top-quark event (Run 349114, Event 1280053930) with seven jets (four of them are b-tagged); two of the top quarks decay leptonically (one with a resulting muon, shown in red, and one with an electron, shown in green), and two top quarks decay hadronically. Green rectangles correspond to energy deposits in cells of the electromagnetic calorimeter, while yellow rectangles correspond to energy deposits in cells of the hadron calorimeter. The jets (b-tagged jets) are shown as yellow (blue) cones. The direction of the missing transverse momentum is indicated by a dotted line.

png (2MB) pdf (2MB)

Tables

Table 01:
Summary of the control regions used in the template fit. M_ee@CV (M_ee@PV) is defined as the invariant mass of the system formed by the track associated with the electron and the closest track at the conversion (primary) vertex. H_T is defined as the scalar sum of the transverse momentum of the isolated leptons and jets.

png (32kB) pdf (57kB)

Table 02:
Normalisation factors for various backgrounds determined from the fit to the control regions. The uncertainties include both statistical and systematic uncertainties.

png (9kB) pdf (39kB)

Table 03:
Post-fit background and signal yields in the full signal region and requiring in addition the BDT score to be greater than 0. The total systematic uncertainty differs from the sum in quadrature of the different uncertainties due to correlations. Q mis-id refers to the charge misassignment background. Mat. Conv. and Low M_ee refer respectively to events with one non-prompt electron originating from photon conversion taking place in the detector material and to events with a virtual photon leading to an e⁺e^- pair. HF e (HF μ) refers to events with one non-prompt electron (muon) from heavy-flavour decay, and LF refers to events with a lepton originating from light-meson decay.

png (29kB) pdf (39kB)

Table 04:
List of the uncertainties on the signal strength μ, grouped in categories.

png (57kB) pdf (42kB)

© 2020 CERN for the benefit of the ATLAS Collaboration. Reproduction of the article, figures and tables on this page is allowed as specified in the CC-BY-4.0 license.
2024-04-26 01:23:47