Observed two-dimensional likelihood scans for $c_{Hq}^{(1)}$ vs. $c_{Hq}^{(3)}$ while other coefficients are fixed at their SM values.

Observed two-dimensional likelihood scans for $c_{Hq}^{(1)}$ vs. $c_{Hu}$ while allowing the other coefficients to float freely at each point of the sca.

Observed two-dimensional likelihood scans for $c_{Hq}^{(1)}$ vs. $c_{Hu}$ while other coefficients are fixed at their SM values.

Observed two-dimensional likelihood scans for $c_{Hq}^{(1)}$ vs. $c_{Hd}$ while allowing the other coefficients to float freely at each point of the sca.

Observed two-dimensional likelihood scans for $c_{Hq}^{(1)}$ vs. $c_{Hd}$ while other coefficients are fixed at their SM values.

Observed two-dimensional likelihood scans for $c_{Hq}^{(3)}$ vs. $c_{Hu}$ while allowing the other coefficients to float freely at each point of the sca.

Observed two-dimensional likelihood scans for $c_{Hq}^{(3)}$ vs. $c_{Hu}$ while other coefficients are fixed at their SM values.

Observed two-dimensional likelihood scans for $c_{Hq}^{(3)}$ vs. $c_{Hd}$ while allowing the other coefficients to float freely at each point of the sca.

Observed two-dimensional likelihood scans for $c_{Hq}^{(3)}$ vs. $c_{Hd}$ while other coefficients are fixed at their SM values.

Observed two-dimensional likelihood scans for $c_{Hu}$ vs. $c_{Hd}$ while allowing the other coefficients to float freely at each point of the sca.

Observed two-dimensional likelihood scans for $c_{Hu}$ vs. $c_{Hd}$ while other coefficients are fixed at their SM values.

Observed two-dimensional likelihood scans for $c_{Hq}^{(1)}$ vs. $g^{(2)}_{ZZ}$ while allowing the other coefficients to float freely at each point of the sca.

Observed two-dimensional likelihood scans for $c_{Hq}^{(1)}$ vs. $g^{(2)}_{ZZ}$ while other coefficients are fixed at their SM values.

Observed two-dimensional likelihood scans for $c_{Hq}^{(3)}$ vs. $g^{(2)}_{ZZ}$ while allowing the other coefficients to float freely at each point of the sca.

Observed two-dimensional likelihood scans for $c_{Hq}^{(3)}$ vs. $g^{(2)}_{ZZ}$ while other coefficients are fixed at their SM values.

Observed two-dimensional likelihood scans for $c_{Hu}$ vs. $g^{(2)}_{ZZ}$ while allowing the other coefficients to float freely at each point of the sca.

Observed two-dimensional likelihood scans for $c_{Hu}$ vs. $g^{(2)}_{ZZ}$ while other coefficients are fixed at their SM values.

Observed two-dimensional likelihood scans for $c_{Hq}^{(1)}$ vs. $g^{(4)}_{ZZ}$ while allowing the other coefficients to float freely at each point of the sca.

Observed two-dimensional likelihood scans for $c_{Hq}^{(1)}$ vs. $g^{(4)}_{ZZ}$ while other coefficients are fixed at their SM values.

Observed two-dimensional likelihood scans for $c_{Hq}^{(3)}$ vs. $g^{(4)}_{ZZ}$ while allowing the other coefficients to float freely at each point of the sca.

Observed two-dimensional likelihood scans for $c_{Hq}^{(3)}$ vs. $g^{(4)}_{ZZ}$ while other coefficients are fixed at their SM values.

Observed two-dimensional likelihood scans for $c_{Hu}$ vs. $g^{(4)}_{ZZ}$ while allowing the other coefficients to float freely at each point of the sca.

Observed two-dimensional likelihood scans for $c_{Hu}$ vs. $g^{(4)}_{ZZ}$ while other coefficients are fixed at their SM values.

Observed two-dimensional likelihood scans for $c_{Hd}$ vs. $g^{(2)}_{ZZ}$ while allowing the other coefficients to float freely at each point of the sca.

Observed two-dimensional likelihood scans for $c_{Hd}$ vs. $g^{(2)}_{ZZ}$ while other coefficients are fixed at their SM values.

Observed two-dimensional likelihood scans for $c_{Hd}$ vs. $g^{(4)}_{ZZ}$ while allowing the other coefficients to float freely at each point of the sca.

Observed two-dimensional likelihood scans for $c_{Hd}$ vs. $g^{(4)}_{ZZ}$ while other coefficients are fixed at their SM values.

Observed two-dimensional likelihood scans for $g^{(2)}_{ZZ}$ vs. $g^{(4)}_{ZZ}$ while allowing the other coefficients to float freely at each point of the sca.

Observed two-dimensional likelihood scans for $g^{(2)}_{ZZ}$ vs. $g^{(4)}_{ZZ}$ while other coefficients are fixed at their SM values.

Tuned intrinsic kT parameters ShowerHandler:IntrinsicPtGaussian in Herwig with the underlying-event tune CH2 at nucleon-nucleon center-of-mass energy from 38.8 GeV to 13 TeV.

Tuned intrinsic kT parameters ShowerHandler:IntrinsicPtGaussian in Herwig with the underlying-event tune CH3 at nucleon-nucleon center-of-mass energy from 38.8 GeV to 13 TeV.

Tuned intrinsic kT parameters BeamRemnants:PrimordialkThard in Pythia with the underlying-event tune CP5 and ISR starting scale SpaceShower:pT0Ref = 1 GeV at nucleon-nucleon center-of-mass energy from 38.8 GeV to 13 TeV.

Tuned intrinsic kT parameters ShowerHandler:IntrinsicPtGaussian in Herwig with the underlying-event tune CH3 and ISR starting scale SudakovCommon:pTmin=0.7 GeV at nucleon-nucleon center-of-mass energy from 38.8 GeV to 13 TeV.

Tuned intrinsic kT parameters BeamRemnants:PrimordialkThard in Pythia with the underlying-event tune CP5 at proton-proton center-of-mass energy 38.8 GeV versus the dilepton mass range of the process.

Tuned intrinsic kT parameters BeamRemnants:PrimordialkThard in Pythia with the underlying-event tune CP5 at proton-proton center-of-mass energy 8000 GeV versus the dilepton mass range of the process.

Tuned intrinsic kT parameters BeamRemnants:PrimordialkThard in Pythia with the underlying-event tune CP5 at proton-nucleon center-of-mass energy 8160 GeV versus the dilepton mass range of the process.

Tuned intrinsic kT parameters BeamRemnants:PrimordialkThard in Pythia with the underlying-event tune CP5 at proton-nucleon center-of-mass energy 13000 GeV versus the dilepton mass range of the process.

Tuned intrinsic kT parameters ShowerHandler:IntrinsicPtGaussian in Herwig with the underlying-event tune CH2 at proton-proton center-of-mass energy 38.8 GeV versus the dilepton mass range of the process.

Tuned intrinsic kT parameters ShowerHandler:IntrinsicPtGaussian in Herwig with the underlying-event tune CH2 at proton-proton center-of-mass energy 8000 GeV versus the dilepton mass range of the process.

Tuned intrinsic kT parameters ShowerHandler:IntrinsicPtGaussian in Herwig with the underlying-event tune CH2 at proton-nucleon center-of-mass energy 8160 GeV versus the dilepton mass range of the process.

Tuned intrinsic kT parameters ShowerHandler:IntrinsicPtGaussian in Herwig with the underlying-event tune CH2 at proton-proton center-of-mass energy 13000 GeV versus the dilepton mass range of the process.

Covariance of the data uncertainty contribution to the tuned intrinsic kT parameters among various generator settings at proton-proton center-of-mass energy 38.8 GeV.

Covariance of the data uncertainty contribution to the tuned intrinsic kT parameters among various generator settings at proton-proton center-of-mass energy 62 GeV.

Covariance of the data uncertainty contribution to the tuned intrinsic kT parameters among various generator settings at proton-proton center-of-mass energy 200 GeV.

Covariance of the data uncertainty contribution to the tuned intrinsic kT parameters among various generator settings at proton-anti-proton center-of-mass energy 1800 GeV.

Covariance of the data uncertainty contribution to the tuned intrinsic kT parameters among various generator settings at proton-anti-proton center-of-mass energy 1960 GeV.

Covariance of the data uncertainty contribution to the tuned intrinsic kT parameters among various generator settings at proton-proton center-of-mass energy 2760 GeV.

Covariance of the data uncertainty contribution to the tuned intrinsic kT parameters among various generator settings at proton-proton center-of-mass energy 8000 GeV.

Covariance of the data uncertainty contribution to the tuned intrinsic kT parameters among various generator settings at proton-nucleon center-of-mass energy 8160 GeV.

Covariance of the data uncertainty contribution to the tuned intrinsic kT parameters among various generator settings at proton-proton center-of-mass energy 13000 GeV when fitting to the CMS measurement.

Covariance of the data uncertainty contribution to the tuned intrinsic kT parameters among various generator settings at proton-proton center-of-mass energy 13000 GeV when fitting to the LHCb measurement.

Covariance of the data uncertainty contribution to the tuned intrinsic kT parameters between the two generator settings of Pythia at proton-proton center-of-mass energy 38.8 GeV.

Covariance of the data uncertainty contribution to the tuned intrinsic kT parameters between the two generator settings of Pythia at proton-proton center-of-mass energy 62 GeV.

Covariance of the data uncertainty contribution to the tuned intrinsic kT parameters between the two generator settings of Pythia at proton-proton center-of-mass energy 200 GeV.

Covariance of the data uncertainty contribution to the tuned intrinsic kT parameters between the two generator settings of Pythia at proton-anti-proton center-of-mass energy 1800 GeV.

Covariance of the data uncertainty contribution to the tuned intrinsic kT parameters between the two generator settings of Pythia at proton-anti-proton center-of-mass energy 1960 GeV.

Covariance of the data uncertainty contribution to the tuned intrinsic kT parameters between the two generator settings of Pythia at proton-proton center-of-mass energy 2760 GeV.

Covariance of the data uncertainty contribution to the tuned intrinsic kT parameters between the two generator settings of Pythia at proton-proton center-of-mass energy 8000 GeV.

Covariance of the data uncertainty contribution to the tuned intrinsic kT parameters between the two generator settings of Pythia at proton-nucleon center-of-mass energy 8160 GeV.

Covariance of the data uncertainty contribution to the tuned intrinsic kT parameters between the two generator settings of Pythia at proton-proton center-of-mass energy 13000 GeV when fitting to the CMS measurement.

Covariance of the data uncertainty contribution to the tuned intrinsic kT parameters between the two generator settings of Pythia at proton-proton center-of-mass energy 13000 GeV when fitting to the LHCb measurement.

Covariance of the data uncertainty contribution to the tuned intrinsic kT parameters between the two generator settings of Herwig at proton-proton center-of-mass energy 38.8 GeV.

Covariance of the data uncertainty contribution to the tuned intrinsic kT parameters between the two generator settings of Herwig at proton-proton center-of-mass energy 62 GeV.

Covariance of the data uncertainty contribution to the tuned intrinsic kT parameters between the two generator settings of Herwig at proton-proton center-of-mass energy 200 GeV.

Covariance of the data uncertainty contribution to the tuned intrinsic kT parameters between the two generator settings of Herwig at proton-anti-proton center-of-mass energy 1800 GeV.

Covariance of the data uncertainty contribution to the tuned intrinsic kT parameters between the two generator settings of Herwig at proton-anti-proton center-of-mass energy 1960 GeV.

Covariance of the data uncertainty contribution to the tuned intrinsic kT parameters between the two generator settings of Herwig at proton-proton center-of-mass energy 2760 GeV.

Covariance of the data uncertainty contribution to the tuned intrinsic kT parameters between the two generator settings of Herwig at proton-proton center-of-mass energy 8000 GeV.

Covariance of the data uncertainty contribution to the tuned intrinsic kT parameters between the two generator settings of Herwig at proton-nucleon center-of-mass energy 8160 GeV.

Covariance of the data uncertainty contribution to the tuned intrinsic kT parameters between the two generator settings of Herwig at proton-proton center-of-mass energy 13000 GeV when fitting to the CMS measurement.

Covariance of the data uncertainty contribution to the tuned intrinsic kT parameters between the two generator settings of Herwig at proton-proton center-of-mass energy 13000 GeV when fitting to the LHCb measurement.

The model-independent 95\% \CL expected and observed upper limits set on ${\sigma(\PP\to 2\Pa+\PX)\mathcal{B}^2(\Pa\to 2\PGm)\alphaGen}$ over the range $0.21 < \MPa < 9\GeV$ for the combined 2016, 2017, and 2018 analyses. Mass ranges that overlap with \JPsi and \PgU resonances are excluded from the search

The 95\% \CL observed upper limits on the effective coupling $\CahLambda$ of the ALP to the SM Higgs assuming the branching fraction of the ALP to muons is 1 (blue) and 0.1 (orange), for both the expected (dashed) and observed (solid) limits

The 95\% \CL observed upper limits on the effective coupling $\cll/\Lambda$ of the ALP to the SM leptons. The orange shaded region represents the parameter space excluded by this search under three choices of $\CahLambda$ $1\TeV^{-2}$ (solid), $0.1\TeV^{-2}$ (dashed), and $0.01\TeV^{-2}$ (dotted)

The 95\% \CL observed upper limits on $\varepsilon^{2} \mathcal{B}(\ZDark \to \sDark \overline{\sDark}) \mathcal{B}^{2}(\sDark \rightarrow 2\PGm)$ for the vector portal model as a function of the dark scalar mass $\MsDark$ and dark vector boson mass $\MZDark$. The yellow and green bands indicate one and two standard deviation values of the limit, respectively. Because the model-independent limits are calculated only up to a dimuon mass of 60\GeV, the \sDark mass considered for these limits is below 60\GeV

The 95\% \CL observed upper limits for $\sigma(\PP\to\PhOneTwo\to 2\PaOne) \mathcal{B}^2(\PaOne\to 2\PGm)$ for the NMSSM as a function of $\MPaOne$ for $\MPhOneTwo$ = 90~GEV. The data presented here reflect the results of the 2017 and 2018 datasets combined with the previously published results of the 2016 dataset in \cite{CMS:2018jid}

The 95\% \CL observed upper limits for $\sigma(\PP\to\PhOneTwo\to 2\PaOne) \mathcal{B}^2(\PaOne\to 2\PGm)$ for the NMSSM as a function of $\MPaOne$ for $\MPhOneTwo$ = 125~GEV. The data presented here reflect the results of the 2017 and 2018 datasets combined with the previously published results of the 2016 dataset in \cite{CMS:2018jid}

The 95\% \CL observed upper limits for $\sigma(\PP\to\PhOneTwo\to 2\PaOne) \mathcal{B}^2(\PaOne\to 2\PGm)$ for the NMSSM as a function of $\MPaOne$ for $\MPhOneTwo$ = 150~GEV. The data presented here reflect the results of the 2017 and 2018 datasets combined with the previously published results of the 2016 dataset in \cite{CMS:2018jid}

The 95\% \CL observed upper limits (black solid curves) from this search as interpreted in the dark SUSY scenario for the process $\PP \to \Ph \to 2\nOne \to 2\gammaDark + 2\nDark \to 4\PGm + \PX$, with \mbox{$\MnOne = 60\GeV$} and $\MnDark = 1\GeV$. The limits are presented in the plane of $\varepsilon$ and $\MgammaDark$. The color gradient represents different branching fraction assumptions for \mbox{$\mathcal{B}(\Ph \to 2\nOne \to 2\gammaDark + 2\nDark)$}, ranging from dark orange (0.05\%) to light orange (10\%). The degradation of the limit around 1\GeV is attributed to the drop in the dimuon branching fraction $\mathcal{B}(\gammaDark \to 2\mu)$ due to the dimuon resonance of the $\phi$ meson~\cite{Batell:2009yf}

non prompt $\psi(2S)$ $\lambda_\theta$

Left panel.The vertical bars (boxes) represent the statistical (bin-to-bin systematic) uncertainties, while the horizontal bars give the bin widths. The global uncertainty (of 5.7%) is not graphically represented. The blue line represents the average of all the points. $ 12 < \mathrm{B} \, p_T < 70$ GeV and $ 0 < |y| < 2.4 $. Global uncertanties are not included in the table (5.7%)

Right panel.The vertical bars (boxes) represent the statistical (bin-to-bin systematic) uncertainties, while the horizontal bars give the bin widths. The global uncertainty (of 5.7%) is not graphically represented. The blue line represents the average of all the points. $ 12 < \mathrm{B} \, p_T < 70$ GeV and $ 0 < |y| < 2.4 $. Global uncertanties are not included in the table (5.7%)

The covariance matrix containing the experimental uncertainties of the particle-level $\mathrm{t}\overline{\mathrm{t}}$ differential cross section as a function of the XCone-jet mass.

The covariance matrix containing the model uncertainties of the particle-level $\mathrm{t}\overline{\mathrm{t}}$ differential cross section as a function of the XCone-jet mass.

The normalized particle-level $\mathrm{t}\overline{\mathrm{t}}$ differential cross section in the fiducial region as a function of the XCone-jet mass.

Correlations between bins in the normalized particle-level $\mathrm{t}\overline{\mathrm{t}}$ differential cross section as a function of the XCone-jet mass.

The covariance matrix containing the statistical uncertainties of the normalized particle-level $\mathrm{t}\overline{\mathrm{t}}$ differential cross section as a function of the XCone-jet mass.

The covariance matrix containing the experimental uncertainties of the normalized particle-level $\mathrm{t}\overline{\mathrm{t}}$ differential cross section as a function of the XCone-jet mass.

The covariance matrix containing the model uncertainties of the normalized particle-level $\mathrm{t}\overline{\mathrm{t}}$ differential cross section as a function of the XCone-jet mass.

Relative experimental uncertainties of the particle-level $\mathrm{t}\overline{\mathrm{t}}$ differential cross section in the fiducial region as a function of the XCone-jet mass.

Relative model uncertainties of the particle-level $\mathrm{t}\overline{\mathrm{t}}$ differential cross section in the fiducial region as a function of the XCone-jet mass.

Relative experimental uncertainties of the normalized particle-level $\mathrm{t}\overline{\mathrm{t}}$ differential cross section in the fiducial region as a function of the XCone-jet mass.

Relative model uncertainties of the normalized particle-level $\mathrm{t}\overline{\mathrm{t}}$ differential cross section in the fiducial region as a function of the XCone-jet mass.

Jet multiplicity measured for a leading-pT jet ($p_{T1}$) with 400 < $p_{T1}$ < 800 GeV and for an azimuthal separation between the two leading jets of $0 < \Delta\Phi_{1,2} < 150^{\circ}$. The full breakdown of the uncertainties is displayed, with PU corresponding to Pileup, PREF to Trigger Prefering, PTHAT to the hard-scale (renormalization and factorization scales), MISS and FAKE to the inefficienties and background, LUMI to integrated luminosity. With JES, JER and stat. unc. following the notation in the paper.

Jet multiplicity measured for a leading-pT jet ($p_{T1}$) with 400 < $p_{T1}$ < 800 GeV and for an azimuthal separation between the two leading jets of $150 < \Delta\Phi_{1,2} < 170^{\circ}$. The full breakdown of the uncertainties is displayed, with PU corresponding to Pileup, PREF to Trigger Prefering, PTHAT to the hard-scale (renormalization and factorization scales), MISS and FAKE to the inefficienties and background, LUMI to integrated luminosity. With JES, JER and stat. unc. following the notation in the paper.

Jet multiplicity measured for a leading-pT jet ($p_{T1}$) with 400 < $p_{T1}$ < 800 GeV and for an azimuthal separation between the two leading jets of $170 < \Delta\Phi_{1,2} < 180^{\circ}$. The full breakdown of the uncertainties is displayed, with PU corresponding to Pileup, PREF to Trigger Prefering, PTHAT to the hard-scale (renormalization and factorization scales), MISS and FAKE to the inefficienties and background, LUMI to integrated luminosity. With JES, JER and stat. unc. following the notation in the paper.

Jet multiplicity measured for a leading-pT jet ($p_{T1}$) with $p_{T1}$ > 800 and for an azimuthal separation between the two leading jets of $0 < \Delta\Phi_{1,2} < 150^{\circ}$. The full breakdown of the uncertainties is displayed, with PU corresponding to Pileup, PREF to Trigger Prefering, PTHAT to the hard-scale (renormalization and factorization scales), MISS and FAKE to the inefficienties and background, LUMI to integrated luminosity. With JES, JER and stat. unc. following the notation in the paper.

Jet multiplicity measured for a leading-pT jet ($p_{T1}$) with $p_{T1}$ > 800 and for an azimuthal separation between the two leading jets of $150 < \Delta\Phi_{1,2} < 170^{\circ}$. The full breakdown of the uncertainties is displayed, with PU corresponding to Pileup, PREF to Trigger Prefering, PTHAT to the hard-scale (renormalization and factorization scales), MISS and FAKE to the inefficienties and background, LUMI to integrated luminosity. With JES, JER and stat. unc. following the notation in the paper.

Jet multiplicity measured for a leading-pT jet ($p_{T1}$) with $p_{T1}$ > 800 and for an azimuthal separation between the two leading jets of $170 < \Delta\Phi_{1,2} < 180^{\circ}$. The full breakdown of the uncertainties is displayed, with PU corresponding to Pileup, PREF to Trigger Prefering, PTHAT to the hard-scale (renormalization and factorization scales), MISS and FAKE to the inefficienties and background, LUMI to integrated luminosity. With JES, JER and stat. unc. following the notation in the paper.

Measured transverse momentum of the leading $p_{T}$ jet ($p_{T1}$). The full breakdown of the uncertainties is displayed, with PU corresponding to Pileup, PREF to Trigger Prefering, PTHAT to the hard-scale (renormalization and factorization scales), MISS and FAKE to the inefficienties and background, LUMI to integrated luminosity. With JES, JER and stat. unc. following the notation in the paper.

Measured transverse momentum of the subleading $p_{T}$ jet ($p_{T2}$). The full breakdown of the uncertainties is displayed, with PU corresponding to Pileup, PREF to Trigger Prefering, PTHAT to the hard-scale (renormalization and factorization scales), MISS and FAKE to the inefficienties and background, LUMI to integrated luminosity. With JES, JER and stat. unc. following the notation in the paper.

Measured transverse momentum of the third leading $p_{T}$ jet ($p_{T3}$). The full breakdown of the uncertainties is displayed, with PU corresponding to Pileup, PREF to Trigger Prefering, PTHAT to the hard-scale (renormalization and factorization scales), MISS and FAKE to the inefficienties and background, LUMI to integrated luminosity. With JES, JER and stat. unc. following the notation in the paper.

Measured transverse momentum of the fourth leading $p_{T}$ jet ($p_{T4}$). The full breakdown of the uncertainties is displayed, with PU corresponding to Pileup, PREF to Trigger Prefering, PTHAT to the hard-scale (renormalization and factorization scales), MISS and FAKE to the inefficienties and background, LUMI to integrated luminosity. With JES, JER and stat. unc. following the notation in the paper.

Jet multiplicity distribution in TUnfold binning ($N_{jets},\Delta\phi_{1,2},p_{T1}$) as indicated in the XML file provided as additional resource. The uncertainties follow the notation of Table 1.

Correlation matrix at particle level for the measured jet multiplicity in TUnfold binning ($N_{jets},\Delta\phi_{1,2},p_{T1}$) as indicated in the XML file provided as additional resource.

Jet $p_{T}$ distributions in TUnfold binning ($p_{T1},p_{T2},p_{T3},p_{T4}$) as indicated in the XML file provided as additional resource. The uncertainties follow the notation of Table 1.

Correlation matrix at particle level for the measured jet $p_{T}$ distributions in TUnfold binning ($p_{T1},p_{T2},p_{T3},p_{T4}$) as indicated in the XML file provided as additional resource.