${p}_{T}^{miss}$ distribution of data and MC events in the signal region with the signal stacked on top of the background prediction for a mass hypothesis of ${m}_{stop} = 225GeV$ and ${m}_{LSP} = 50 GeV$. Events from ttW, ttZ, DY, non-prompt leptons, and diboson processes are grouped into the 'Other' category. The lower panel contains the data-to-prediction ratio. The uncertainty band includes statistical, background normalization and all systematic uncertainties.

${p}_{T}^{miss}$ normalized distribution for signal points with different top squark and neutralino masses and SM ttbar events are compared.

${m}_{T2} (e\mu)$ normalized distribution for signal points with different top squark and neutralino masses and SM ttbar events are compared.

$\Delta\eta (e\mu)$ normalized distribution for signal points with different top squark and neutralino masses and SM ttbar events are compared.

$\Delta\phi (e\mu)$ normalized distribution for signal points with different top squark and neutralino masses and SM ttbar events are compared.

Normalized DNN score distribution in the signal region for two mass hypotheses to compare signal and ttbar background: $m_{LSP}=$ 50 (100) GeV and $m_{stop}=$ 225 (275) GeV.

Post-fit DNN score distributions in the signal region with the signal superimposed and scaled by a factor of 20 for different mass hypotheses of $({m}_{stop}, {m}_{LSP})=$ (225, 50) GeV. The uncertainty band includes statistical, background normalization, and all systematic uncertainties. Events from ttW, ttZ, DY, non-prompt leptons, and diboson processes are grouped into the 'Other' category. The lower panel contains the data-to-prediction ratio before the fit (line) and after (dots), each of them with their corresponding band of uncertainties, and the ratio between the sum of the signal and background predictions and the background prediction. The masses of the signal model correspond to the values of the DNN mass parameters in each distribution.

Post-fit DNN score distributions in the signal region with the signal superimposed and scaled by a factor of 20 for different mass hypotheses of $({m}_{stop}, {m}_{LSP})=$ (275, 100) GeV. The uncertainty band includes statistical, background normalization, and all systematic uncertainties. Events from ttW, ttZ, DY, non-prompt leptons, and diboson processes are grouped into the 'Other' category. The lower panel contains the data-to-prediction ratio before the fit (line) and after (dots), each of them with their corresponding band of uncertainties, and the ratio between the sum of the signal and background predictions and the background prediction. The masses of the signal model correspond to the values of the DNN mass parameters in each distribution.

Post-fit DNN score distributions in the signal region with the signal superimposed and scaled by a factor of 20 for different mass hypotheses of $({m}_{stop}, {m}_{LSP})=$ (275, 70) GeV. The uncertainty band includes statistical, background normalization, and all systematic uncertainties. Events from ttW, ttZ, DY, non-prompt leptons, and diboson processes are grouped into the 'Other' category. The lower panel contains the data-to-prediction ratio before the fit (line) and after (dots), each of them with their corresponding band of uncertainties, and the ratio between the sum of the signal and background predictions and the background prediction. The masses of the signal model correspond to the values of the DNN mass parameters in each distribution.

Post-fit DNN score distributions in the signal region with the signal superimposed and scaled by a factor of 20 for different mass hypotheses of $({m}_{stop}, {m}_{LSP})=$ (245, 100) GeV. The uncertainty band includes statistical, background normalization, and all systematic uncertainties. Events from ttW, ttZ, DY, non-prompt leptons, and diboson processes are grouped into the 'Other' category. The lower panel contains the data-to-prediction ratio before the fit (line) and after (dots), each of them with their corresponding band of uncertainties, and the ratio between the sum of the signal and background predictions and the background prediction. The masses of the signal model correspond to the values of the DNN mass parameters in each distribution.

Upper limit at 95% CL on the signal cross section as a function of the top squark and neutralino masses. The model is excluded for all of the colored region inside the black boundary.

Upper limit at 95% CL on the signal cross section as a function of the top squark mass for mass difference of 175 GeV. The green and yellow bands represent the regions containing 68 and 95%, respectively, of the distribution of limits expected under the background-only hypothesis. The purple dotted line indicates the approximate NNLO + NNLL productioncross section.

Upper limit at 95% CL on the signal cross section as a function of the top squark mass for mass difference of 185 GeV. The green and yellow bands represent the regions containing 68 and 95%, respectively, of the distribution of limits expected under the background-only hypothesis. The purple dotted line indicates the approximate NNLO + NNLL productioncross section.

Upper limit at 95% CL on the signal cross section as a function of the top squark mass for mass difference of 165 GeV. The green and yellow bands represent the regions containing 68 and 95%, respectively, of the distribution of limits expected under the background-only hypothesis. The purple dotted line indicates the approximate NNLO + NNLL productioncross section.

Summary of the contributions of the experimental uncertainties in the DNN score distribution for signal and the ttbar background. The values represent the relative variation in the number of expected events across different signal models in the signal region.

Summary of the contribution of each modeling uncertainty source to the DNN score distribution for the ttbar background.

95% CL expected upper limit on top squark cross section for the direct decay model.

95% CL observed upper limit on top squark cross section for the direct decay model.

Contour of the region excluded with 95% CL (expected).

Contour of the region excluded with 95% CL (observed).

95% CL expected upper limit on top squark cross section for the cascade decay model.

95% CL observed upper limit on top squark cross section for the cascade decay model.

Contour of the region excluded with 95% CL (expected).

Contour of the region excluded with 95% CL (observed).

95% CL expected upper limit on top squark cross section for the mixed decay model.

95% CL observed upper limit on top squark cross section for the mixed decay model.

Contour of the region excluded with 95% CL (expected).

Contour of the region excluded with 95% CL (observed).

95% CL expected limit on $\sigma/\sigma_{theory}$ for a fermionic DM particle with m = 1 GeV, as a function of the mediator mass for a scalar.

95% CL observed limit on $\sigma/\sigma_{theory}$ for a fermionic DM particle with m = 1 GeV, as a function of the mediator mass for a scalar.

95% CL expected limit on $\sigma/\sigma_{theory}$ for a fermionic DM particle with m = 1 GeV, as a function of the mediator mass for a pseudoscalar.

95% CL observed limit on $\sigma/\sigma_{theory}$ for a fermionic DM particle with m = 1 GeV, as a function of the mediator mass for a pseudoscalar.

$v_2\{2, |\Delta\eta| > 2.0\}$ of $\pi^{\pm}$ as a function of $p_{\rm T}$ for the 20-30% centrality interval.

$v_2\{2, |\Delta\eta| > 2.0\}$ of $\pi^{\pm}$ as a function of $p_{\rm T}$ for the 30-40% centrality interval.

$v_2\{2, |\Delta\eta| > 2.0\}$ of $\pi^{\pm}$ as a function of $p_{\rm T}$ for the 40-50% centrality interval.

$v_2\{2, |\Delta\eta| > 2.0\}$ of $\pi^{\pm}$ as a function of $p_{\rm T}$ for the 50-60% centrality interval.

$v_2\{2, |\Delta\eta| > 2.0\}$ of $\rm{K}^{\pm}$ as a function of $p_{\rm T}$ for the 0-5% centrality interval.

$v_2\{2, |\Delta\eta| > 2.0\}$ of $\rm{K}^{\pm}$ as a function of $p_{\rm T}$ for the 5-10% centrality interval.

$v_2\{2, |\Delta\eta| > 2.0\}$ of $\rm{K}^{\pm}$ as a function of $p_{\rm T}$ for the 10-20% centrality interval.

$v_2\{2, |\Delta\eta| > 2.0\}$ of $\rm{K}^{\pm}$ as a function of $p_{\rm T}$ for the 20-30% centrality interval.

$v_2\{2, |\Delta\eta| > 2.0\}$ of $\rm{K}^{\pm}$ as a function of $p_{\rm T}$ for the 30-40% centrality interval.

$v_2\{2, |\Delta\eta| > 2.0\}$ of $\rm{K}^{\pm}$ as a function of $p_{\rm T}$ for the 40-50% centrality interval.

$v_2\{2, |\Delta\eta| > 2.0\}$ of $\rm{K}^{\pm}$ as a function of $p_{\rm T}$ for the 50-60% centrality interval.

$v_2\{2, |\Delta\eta| > 2.0\}$ of p+$\rm\overline{p}$ as a function of $p_{\rm T}$ for the 0-5% centrality interval.

$v_2\{2, |\Delta\eta| > 2.0\}$ of p+$\rm\overline{p}$ as a function of $p_{\rm T}$ for the 5-10% centrality interval.

$v_2\{2, |\Delta\eta| > 2.0\}$ of p+$\rm\overline{p}$ as a function of $p_{\rm T}$ for the 10-20% centrality interval.

$v_2\{2, |\Delta\eta| > 2.0\}$ of p+$\rm\overline{p}$ as a function of $p_{\rm T}$ for the 20-30% centrality interval.

$v_2\{2, |\Delta\eta| > 2.0\}$ of p+$\rm\overline{p}$ as a function of $p_{\rm T}$ for the 30-40% centrality interval.

$v_2\{2, |\Delta\eta| > 2.0\}$ of p+$\rm\overline{p}$ as a function of $p_{\rm T}$ for the 40-50% centrality interval.

$v_2\{2, |\Delta\eta| > 2.0\}$ of p+$\rm\overline{p}$ as a function of $p_{\rm T}$ for the 50-60% centrality interval.

$v_2\{2, |\Delta\eta| > 2.0\}$ of ${\rm K}^{0}_{\rm{S}}$ as a function of $p_{\rm T}$ for the 0-5% centrality interval.

$v_2\{2, |\Delta\eta| > 2.0\}$ of ${\rm K}^{0}_{\rm{S}}$ as a function of $p_{\rm T}$ for the 5-10% centrality interval.

$v_2\{2, |\Delta\eta| > 2.0\}$ of ${\rm K}^{0}_{\rm{S}}$ as a function of $p_{\rm T}$ for the 10-20% centrality interval.

$v_2\{2, |\Delta\eta| > 2.0\}$ of ${\rm K}^{0}_{\rm{S}}$ as a function of $p_{\rm T}$ for the 20-30% centrality interval.

$v_2\{2, |\Delta\eta| > 2.0\}$ of ${\rm K}^{0}_{\rm{S}}$ as a function of $p_{\rm T}$ for the 30-40% centrality interval.

$v_2\{2, |\Delta\eta| > 2.0\}$ of ${\rm K}^{0}_{\rm{S}}$ as a function of $p_{\rm T}$ for the 40-50% centrality interval.

$v_2\{2, |\Delta\eta| > 2.0\}$ of ${\rm K}^{0}_{\rm{S}}$ as a function of $p_{\rm T}$ for the 50-60% centrality interval.

$v_2\{2, |\Delta\eta| > 2.0\}$ of $\Lambda$+$\overline{\Lambda}$ as a function of $p_{\rm T}$ for the 0-5% centrality interval.

$v_2\{2, |\Delta\eta| > 2.0\}$ of $\Lambda$+$\overline{\Lambda}$ as a function of $p_{\rm T}$ for the 5-10% centrality interval.

$v_2\{2, |\Delta\eta| > 2.0\}$ of $\Lambda$+$\overline{\Lambda}$ as a function of $p_{\rm T}$ for the 10-20% centrality interval.

$v_2\{2, |\Delta\eta| > 2.0\}$ of $\Lambda$+$\overline{\Lambda}$ as a function of $p_{\rm T}$ for the 20-30% centrality interval.

$v_2\{2, |\Delta\eta| > 2.0\}$ of $\Lambda$+$\overline{\Lambda}$ as a function of $p_{\rm T}$ for the 30-40% centrality interval.

$v_2\{2, |\Delta\eta| > 2.0\}$ of $\Lambda$+$\overline{\Lambda}$ as a function of $p_{\rm T}$ for the 40-50% centrality interval.

$v_2\{2, |\Delta\eta| > 2.0\}$ of $\Lambda$+$\overline{\Lambda}$ as a function of $p_{\rm T}$ for the 50-60% centrality interval.

$v_3\{2, |\Delta\eta| > 2.0\}$ of $\pi^{\pm}$ as a function of $p_{\rm T}$ for the 0-10% centrality interval.

$v_3\{2, |\Delta\eta| > 2.0\}$ of $\pi^{\pm}$ as a function of $p_{\rm T}$ for the 10-30% centrality interval.

$v_3\{2, |\Delta\eta| > 2.0\}$ of $\pi^{\pm}$ as a function of $p_{\rm T}$ for the 30-50% centrality interval.

$v_3\{2, |\Delta\eta| > 2.0\}$ of $\rm{K}^{\pm}$ as a function of $p_{\rm T}$ for the 0-10% centrality interval.

$v_3\{2, |\Delta\eta| > 2.0\}$ of $\rm{K}^{\pm}$ as a function of $p_{\rm T}$ for the 10-30% centrality interval.

$v_3\{2, |\Delta\eta| > 2.0\}$ of $\rm{K}^{\pm}$ as a function of $p_{\rm T}$ for the 30-50% centrality interval.

$v_3\{2, |\Delta\eta| > 2.0\}$ of p+$\rm\overline{p}$ as a function of $p_{\rm T}$ for the 0-10% centrality interval.

$v_3\{2, |\Delta\eta| > 2.0\}$ of p+$\rm\overline{p}$ as a function of $p_{\rm T}$ for the 10-30% centrality interval.

$v_3\{2, |\Delta\eta| > 2.0\}$ of p+$\rm\overline{p}$ as a function of $p_{\rm T}$ for the 30-50% centrality interval.

The $p_{\rm T}$ of $v_2^{\mathrm{max}}$ for $\pi^{\pm}$ as a function of centrality.

The $p_{\rm T}$ of $v_2^{\mathrm{max}}$ for p+$\rm\overline{p}$ as a function of centrality.

Pb-remnant side ZN signal normalized to MB value vs. ZN centrality percentile in p-Pb collisions at 8.16 TeV

p-remnant side ZN signal normalized to MB value vs. ZN centrality percentile in p-Pb collisions at 5.02 TeV

p-remnant side ZN signal normalized to MB value vs. ZN centrality percentile in p-Pb collisions at 8.16 TeV

Pb-remnant side ZN signal normalized to MB value vs. average Ncoll in p-Pb collisions at 5.02 TeV

Pb-remnant side ZN signal normalized to MB value vs. average Ncoll in p-Pb collisions at 8.16 TeV

p-remnant side ZN signal normalized to MB value vs. average Ncoll in p-Pb collisions at 5.02 TeV

p-remnant side ZN signal normalized to MB value vs. average Ncoll in p-Pb collisions at 8.16 TeV

ZN signal normalized to MB value vs. normalized charged-particle multiplicity in p-p collisions at 13 TeV

p-remnant side ZN signal normalized to MB value vs. normalized charged-particle multiplicity in p-Pb collisions at 8.16 TeV

ZP signal normalized to MB value vs. normalized charged-particle multiplicity in p-p collisions at 13 TeV

p-remnant side ZP signal normalized to MB value vs. average charged-particle multiplicity in p-Pb collisions at 8.16 TeV

ZN signal normalized to MB value vs. leading transverse momentum in |eta|<0.8 in p-p collisions at 13 TeV

ZP signal normalized to MB value vs. leading transverse momentum in |eta|<0.8 in p-p collisions at 13 TeV

Differential cross section as a function of $m_{\gamma\gamma}$. The table contains the values measured in data and theory predictions from SHERPA, DIPHOX and NNLOJET.

Differential cross section as a function of $p_{T,\gamma\gamma}$. The table contains the values measured in data and theory predictions from SHERPA, DIPHOX and NNLOJET.

Differential cross section as a function of $a_{T,\gamma\gamma}$. The table contains the values measured in data and theory predictions from SHERPA, DIPHOX and NNLOJET.

Differential cross section as a function of $\phi_{\eta}*$. The table contains the values measured in data and theory predictions from SHERPA, DIPHOX and NNLOJET.

Differential cross section as a function of $\pi-\Delta\phi_{\gamma\gamma}$. The table contains the values measured in data and theory predictions from SHERPA, DIPHOX and NNLOJET.

Differential cross section as a function of $|cos\theta*|^{(CS)}$. The table contains the values measured in data and theory predictions from SHERPA, DIPHOX and NNLOJET.

$v_3$ vs $p_T$, p+Au at 200 GeV, 0-5% central, BBCS-FVTXS-CNT detector combination

$v_3$ vs $p_T$, d+Au at 200 GeV, 0-5% central, BBCS-FVTXS-CNT detector combination

$v_3$ vs $p_T$, 3He+Au at 200 GeV, 0-5% central, BBCS-FVTXS-CNT detector combination

$v_2$ vs $p_T$, p+Au at 200 GeV, 0-5% central, FVTXS-CNT-FVTXN detector combination

$v_2$ vs $p_T$, d+Au at 200 GeV, 0-5% central, FVTXS-CNT-FVTXN detector combination

$v_2$ vs $p_T$, 3He+Au at 200 GeV, 0-5% central, FVTXS-CNT-FVTXN detector combination

$v_2$ ratio vs $p_T$, p/d/3He+Au at 200 GeV, 0-5% central

$v_3$ vs $p_T$, p+Au at 200 GeV, 0-5% central, FVTXS-CNT-FVTXN detector combination

$v_3$ vs $p_T$, d+Au at 200 GeV, 0-5% central, FVTXS-CNT-FVTXN detector combination

$v_3$ vs $p_T$, 3He+Au at 200 GeV, 0-5% central, FVTXS-CNT-FVTXN detector combination

$c_2$ vs $\eta$, p+Au at 200 GeV, 0-5% central

$c_2$ vs $\eta$, d+Au at 200 GeV, 0-5% central

$c_2$ vs $\eta$, 3He+Au at 200 GeV, 0-5% central

$c_3$ vs $\eta$, p+Au at 200 GeV, 0-5% central

$c_3$ vs $\eta$, d+Au at 200 GeV, 0-5% central

$c_3$ vs $\eta$, 3He+Au at 200 GeV, 0-5% central

$C(\Delta\phi)$ vs $\Delta\phi$, p+p at 200 GeV, minimum bias, BBCS-FVTXS correlation

$C(\Delta\phi)$ vs $\Delta\phi$, p+Au at 200 GeV, 0-5% central, BBCS-FVTXS correlation

$C(\Delta\phi)$ vs $\Delta\phi$, d+Au at 200 GeV, 0-5% central, BBCS-FVTXS correlation

$C(\Delta\phi)$ vs $\Delta\phi$, 3He+Au at 200 GeV, 0-5% central, BBCS-FVTXS correlation

$C(\Delta\phi)$ vs $\Delta\phi$, p+p at 200 GeV, minimum bias, BBCS-FVTXN correlation

$C(\Delta\phi)$ vs $\Delta\phi$, p+Au at 200 GeV, 0-5% central, BBCS-FVTXN correlation

$C(\Delta\phi)$ vs $\Delta\phi$, d+Au at 200 GeV, 0-5% central, BBCS-FVTXN correlation

$C(\Delta\phi)$ vs $\Delta\phi$, 3He+Au at 200 GeV, 0-5% central, BBCS-FVTXN correlation

$C(\Delta\phi)$ vs $\Delta\phi$, p+p at 200 GeV, minimum bias, FVTXS-FVTXN correlation

$C(\Delta\phi)$ vs $\Delta\phi$, p+Au at 200 GeV, 0-5% central, FVTXS-FVTXN correlation

$C(\Delta\phi)$ vs $\Delta\phi$, d+Au at 200 GeV, 0-5% central, FVTXS-FVTXN correlation

$C(\Delta\phi)$ vs $\Delta\phi$, 3He+Au at 200 GeV, 0-5% central, FVTXS-FVTXN correlation

$C(\Delta\phi)$ vs $\Delta\phi$, p+p at 200 GeV, minimum bias, CNT-BBCS correlation

$C(\Delta\phi)$ vs $\Delta\phi$, p+Au at 200 GeV, 0-5% central, CNT-BBCS correlation

$C(\Delta\phi)$ vs $\Delta\phi$, d+Au at 200 GeV, 0-5% central, CNT-BBCS correlation

$C(\Delta\phi)$ vs $\Delta\phi$, 3He+Au at 200 GeV, 0-5% central, CNT-BBCS correlation

$C(\Delta\phi)$ vs $\Delta\phi$, p+p at 200 GeV, minimum bias, CNT-FVTXS correlation

$C(\Delta\phi)$ vs $\Delta\phi$, p+Au at 200 GeV, 0-5% central, CNT-FVTXS correlation

$C(\Delta\phi)$ vs $\Delta\phi$, d+Au at 200 GeV, 0-5% central, CNT-FVTXS correlation

$C(\Delta\phi)$ vs $\Delta\phi$, 3He+Au at 200 GeV, 0-5% central, CNT-FVTXS correlation

$C(\Delta\phi)$ vs $\Delta\phi$, p+p at 200 GeV, minimum bias, CNT-FVTXN correlation

$C(\Delta\phi)$ vs $\Delta\phi$, p+Au at 200 GeV, 0-5% central, CNT-FVTXN correlation

$C(\Delta\phi)$ vs $\Delta\phi$, d+Au at 200 GeV, 0-5% central, CNT-FVTXN correlation

$C(\Delta\phi)$ vs $\Delta\phi$, 3He+Au at 200 GeV, 0-5% central, CNT-FVTXN correlation

$c_n$ vs detector combination, p+p at 200 GeV, Minimum Bias

$c_n$ vs $p_T$, p+p at 200 GeV, Minimum Bias, BBCS-CNT detector combination

$c_n$ vs $p_T$, p+p at 200 GeV, Minimum Bias, FVTXS-CNT detector combination

$c_n$ vs $p_T$, p+p at 200 GeV, Minimum Bias, FVTXN-CNT detector combination

$c_n$ vs detector combination, p+Au at 200 GeV, 0-5% central

$c_n$ vs $p_T$, p+Au at 200 GeV, 0-5% central, BBCS-CNT detector combination

$c_n$ vs $p_T$, p+Au at 200 GeV, 0-5% central, FVTXS-CNT detector combination

$c_n$ vs $p_T$, p+Au at 200 GeV, 0-5% central, FVTXN-CNT detector combination

$c_n$ vs detector combination, p+Au at 200 GeV, 0-5% central

$c_n$ vs $p_T$, p+Au at 200 GeV, 0-5% central, BBCS-CNT detector combination

$c_n$ vs $p_T$, p+Au at 200 GeV, 0-5% central, FVTXS-CNT detector combination

$c_n$ vs $p_T$, p+Au at 200 GeV, 0-5% central, FVTXN-CNT detector combination

$c_n$ vs detector combination, p+Au at 200 GeV, 0-5% central

$c_n$ vs $p_T$, 3He+Au at 200 GeV, 0-5% central, BBCS-CNT detector combination

$c_n$ vs $p_T$, 3He+Au at 200 GeV, 0-5% central, FVTXS-CNT detector combination

$c_n$ vs $p_T$, 3He+Au at 200 GeV, 0-5% central, FVTXN-CNT detector combination

95% CL exclusion limits on $\mathcal{B}(H\rightarrow aa \rightarrow b\bar{b}b\bar{b})$ for $m_a = 45$ GeV.

95% CL exclusion limits on $\mathcal{B}(H\rightarrow aa \rightarrow b\bar{b}b\bar{b})$ for $m_a = 55$ GeV.

The fraction of $a$ boson decays matched to reconstructed displaced vertices passing all vertex selections in signal MC.

The extrapolated signal selection efficiency as a function of $c\tau_{a}$.

The 95% CL observed and expected limits on the branching fraction B(H $\rightarrow$ SS) for 15 GeV mass and $ S \rightarrow d\bar{d}$ decay mode.

The 95% CL observed and expected limits on the branching fraction B(H $\rightarrow$ SS) for 40 GeV mass and $ S \rightarrow d\bar{d}$ decay mode.

The 95% CL observed and expected limits on the branching fraction B(H $\rightarrow$ SS) for 40 GeV mass and $ S \rightarrow d\bar{d}$ decay mode.

The 95% CL observed and expected limits on the branching fraction B(H $\rightarrow$ SS) for 55 GeV mass and $ S \rightarrow d\bar{d}$ decay mode.

The 95% CL observed and expected limits on the branching fraction B(H $\rightarrow$ SS) for 55 GeV mass and $ S \rightarrow d\bar{d}$ decay mode.

The 95% CL observed and expected limits on the branching fraction B(H $\rightarrow$ SS) for 7 GeV mass and $ S \rightarrow \tau^{+} \tau^{-}$ decay mode.

The 95% CL observed and expected limits on the branching fraction B(H $\rightarrow$ SS) for 7 GeV mass and $ S \rightarrow \tau^{+} \tau^{-}$ decay mode.

The 95% CL observed and expected limits on the branching fraction B(H $\rightarrow$ SS) for 15 GeV mass and $ S \rightarrow \tau^{+} \tau^{-}$ decay mode.

The 95% CL observed and expected limits on the branching fraction B(H $\rightarrow$ SS) for 15 GeV mass and $ S \rightarrow \tau^{+} \tau^{-}$ decay mode.

The 95% CL observed and expected limits on the branching fraction B(H $\rightarrow$ SS) for 40 GeV mass and $ S \rightarrow \tau^{+} \tau^{-}$ decay mode.

The 95% CL observed and expected limits on the branching fraction B(H $\rightarrow$ SS) for 40 GeV mass and $ S \rightarrow \tau^{+} \tau^{-}$ decay mode.

The 95% CL observed and expected limits on the branching fraction B(H $\rightarrow$ SS) for 55 GeV mass and $ S \rightarrow \tau^{+} \tau^{-}$ decay mode.

The 95% CL observed and expected limits on the branching fraction B(H $\rightarrow$ SS) for 55 GeV mass and $ S \rightarrow \tau^{+} \tau^{-}$ decay mode.

The 95% CL observed and expected limits on the branching fraction B(H $\rightarrow$ SS) for 15 GeV mass and $ S \rightarrow b\bar{b}$ decay mode.

The 95% CL observed and expected limits on the branching fraction B(H $\rightarrow$ SS) for 15 GeV mass and $ S \rightarrow b\bar{b}$ decay mode.

The 95% CL observed and expected limits on the branching fraction B(H $\rightarrow$ SS) for 40 GeV mass and $ S \rightarrow b\bar{b}$ decay mode.

The 95% CL observed and expected limits on the branching fraction B(H $\rightarrow$ SS) for 40 GeV mass and $ S \rightarrow b\bar{b}$ decay mode.

The 95% CL observed and expected limits on the branching fraction B(H $\rightarrow$ SS) for 55 GeV mass and $ S \rightarrow b\bar{b}$ decay mode.

The 95% CL observed and expected limits on the branching fraction B(H $\rightarrow$ SS) for 55 GeV mass and $ S \rightarrow b\bar{b}$ decay mode.

The cluster efficiency in bins of hadronic and EM energy in region A. Region A is defined as 391 cm $< r <$ 695.5 cm and 400 cm $< |z| <$ 671 cm. The cluster efficiency is estimated with LLPs decaying to $\tau^{+} \tau^{-}$. The sample contains equal fractions of events with LLP mass of 7, 15, 40, and 55 GeV and LLP lifetime of 0.1, 1, 10, and 100m. The first hadronic energy bins correspond to LLPs that decayed leptonically with 0 hadronic energy. The cluster efficiency includes all cluster-level selections described in the paper, except for the jet veto, time cut, and $\Delta\phi$ cut. The full simulation signal yield prediction for samples with various LLP mass between 7 - 55 GeV, lifetime between 0.1 - 100 m, and decay mode to $d\bar{d}$ and $\tau^{+} \tau^{-}$ can be reproduced using this parameterization to within 35% and 20% for region A and B, respectively.

The cluster efficiency in bins of hadronic and EM energy in region A. Region A is defined as 391 cm $< r <$ 695.5 cm and 400 cm $< |z| <$ 671 cm. The cluster efficiency is estimated with LLPs decaying to $\tau^{+} \tau^{-}$. The sample contains equal fractions of events with LLP mass of 7, 15, 40, and 55 GeV and LLP lifetime of 0.1, 1, 10, and 100m. The first hadronic energy bins correspond to LLPs that decayed leptonically with 0 hadronic energy. The cluster efficiency includes all cluster-level selections described in the paper, except for the jet veto, time cut, and $\Delta\phi$ cut. The full simulation signal yield prediction for samples with various LLP mass between 7 - 55 GeV, lifetime between 0.1 - 100 m, and decay mode to $d\bar{d}$ and $\tau^{+} \tau^{-}$ can be reproduced using this parameterization to within 35% and 20% for region A and B, respectively.

The cluster efficiency in bins of hadronic and EM energy in region B. Region B is defined as 671 cm $< |z| <$ 1100 cm, $r <$ 695.5 cm, and $|\eta| <$ 2. The cluster efficiency is estimated with LLPs decaying to $\tau^{+} \tau^{-}$. The sample contains equal fractions of events with LLP mass of 7, 15, 40, and 55 GeV and LLP lifetime of 0.1, 1, 10, and 100m. The first hadronic energy bins correspond to LLPs that decayed leptonically with 0 hadronic energy. The cluster efficiency includes all cluster-level selections described in the paper, except for the jet veto, time cut, and $\Delta\phi$ cut. The full simulation signal yield prediction for samples with various LLP mass between 7 - 55 GeV, lifetime between 0.1 - 100 m, and decay mode to $d\bar{d}$ and $\tau^{+} \tau^{-}$ can be reproduced using this parameterization to within 35% and 20% for region A and B, respectively.

The cluster efficiency in bins of hadronic and EM energy in region B. Region B is defined as 671 cm $< |z| <$ 1100 cm, $r <$ 695.5 cm, and $|\eta| <$ 2. The cluster efficiency is estimated with LLPs decaying to $\tau^{+} \tau^{-}$. The sample contains equal fractions of events with LLP mass of 7, 15, 40, and 55 GeV and LLP lifetime of 0.1, 1, 10, and 100m. The first hadronic energy bins correspond to LLPs that decayed leptonically with 0 hadronic energy. The cluster efficiency includes all cluster-level selections described in the paper, except for the jet veto, time cut, and $\Delta\phi$ cut. The full simulation signal yield prediction for samples with various LLP mass between 7 - 55 GeV, lifetime between 0.1 - 100 m, and decay mode to $d\bar{d}$ and $\tau^{+} \tau^{-}$ can be reproduced using this parameterization to within 35% and 20% for region A and B, respectively.

The efficiency of $N_{station} > 1$ requirement in bins of hadronic energy in region B. Region B is defined as 671 cm $< |z| <$ 1100 cm, $r <$ 695.5 cm, and $|\eta| <$ 2. The cluster efficiency is estimated with LLPs decaying to $\tau^{+} \tau^{-}$. The sample contains equal fractions of events with LLP mass of 7, 15, 40, and 55 GeV and LLP lifetime of 0.1, 1, 10, and 100m. The first hadronic energy bin corresponds to LLPs that decayed leptonically with 0 hadronic energy. The efficiency is calculated with respect to clusters that pass all cluster-level cuts described in the paper, except for the jet veto, time cut, and $\Delta\phi$ cut. The full simulation signal yield prediction for samples with various LLP mass between 7 - 55 GeV, lifetime between 0.1 - 100 m, and decay mode to $d\bar{d}$ and $\tau^{+} \tau^{-}$ can be reproduced using this parameterization to within 10%.

The efficiency of $N_{station} > 1$ requirement in bins of hadronic energy in region B. Region B is defined as 671 cm $< |z| <$ 1100 cm, $r <$ 695.5 cm, and $|\eta| <$ 2. The cluster efficiency is estimated with LLPs decaying to $\tau^{+} \tau^{-}$. The sample contains equal fractions of events with LLP mass of 7, 15, 40, and 55 GeV and LLP lifetime of 0.1, 1, 10, and 100m. The first hadronic energy bin corresponds to LLPs that decayed leptonically with 0 hadronic energy. The efficiency is calculated with respect to clusters that pass all cluster-level cuts described in the paper, except for the jet veto, time cut, and $\Delta\phi$ cut. The full simulation signal yield prediction for samples with various LLP mass between 7 - 55 GeV, lifetime between 0.1 - 100 m, and decay mode to $d\bar{d}$ and $\tau^{+} \tau^{-}$ can be reproduced using this parameterization to within 10%.

The geometric signal acceptance as a function of $c\tau$. The acceptance is shown for four different LLP mass hypotheses: 7, 15, 40, and 55 GeV. The acceptance is defined by requiring at least 1 LLP to decay in the region defined as 400 cm $< |z| <$ 1100 cm, $r < $ 695.5 cm, and $|\eta| <$ 2.4.

Distribution of $m_{T}(W)$ in the $N_{jet}\geq 3$ signal region.

Distribution of $M_{3}$ in the $N_{jet}\geq 3$ signal region.

Distribution of $M_{3}$ in the $N_{jet}\geq 3$ signal region.

Distribution of $m(l,\gamma)$ in the $N_{jet}\geq 3$ signal region.

Distribution of $m(l,\gamma)$ in the $N_{jet}\geq 3$ signal region.

Distribution of $\Delta R(l,\gamma)$ in the $N_{jet}\geq 3$ signal region.

Distribution of $\Delta R(l,\gamma)$ in the $N_{jet}\geq 3$ signal region.

Distribution of $\Delta R(j,\gamma)$ in the $N_{jet}\geq 3$ signal region.

Distribution of $\Delta R(j,\gamma)$ in the $N_{jet}\geq 3$ signal region.

Fit result of the multijet template obtained with loosely isolated leptons and the electroweak background to the measured $m_{T}(W)$ distribution with isolated leptons in the $N_{jet}=2$, $N_{b jet}=0$ selection for electrons.

Fit result of the multijet template obtained with loosely isolated leptons and the electroweak background to the measured $m_{T}(W)$ distribution with isolated leptons in the $N_{jet}=2$, $N_{b jet}=0$ selection for electrons.

Fit result of the multijet template obtained with loosely isolated leptons and the electroweak background to the measured $m_{T}(W)$ distribution with isolated leptons in the $N_{jet}=2$, $N_{b jet}=0$ selection for muons.

Fit result of the multijet template obtained with loosely isolated leptons and the electroweak background to the measured $m_{T}(W)$ distribution with isolated leptons in the $N_{jet}=2$, $N_{b jet}=0$ selection for muons.

Distribution of the invariant mass of the lepton and the photon ($m(l,\gamma)$) in the $N_{jet}\geq 3$, $N_{b jet}=0$ selection for the e channel.

Distribution of the invariant mass of the lepton and the photon ($m(l,\gamma)$) in the $N_{jet}\geq 3$, $N_{b jet}=0$ selection for the e channel.

Distribution of the invariant mass of the lepton and the photon ($m(l,\gamma)$) in the $N_{jet}\geq 3$, $N_{b jet}=0$ selection for the $\mu$ channel.

Distribution of the invariant mass of the lepton and the photon ($m(l,\gamma)$) in the $N_{jet}\geq 3$, $N_{b jet}=0$ selection for the $\mu$ channel.

Extracted scale factors for the contribution from misidentified electrons for the three data-taking periods, and the Z$\gamma$, W$\gamma$ simulations.

Extracted scale factors for the contribution from misidentified electrons for the three data-taking periods, and the Z$\gamma$, W$\gamma$ simulations.

Predicted and observed yields in the control regions in the $N_{jet}= 3$ and $\geq 4$ seletions using the post-fit values of the nuisance parameters.

Predicted and observed yields in the control regions in the $N_{jet}= 3$ and $\geq 4$ seletions using the post-fit values of the nuisance parameters.

Predicted and observed yields in the signal regions in the $N_{jet}= 3$ and $\geq 4$ seletions using the post-fit values of the nuisance parameters.

Predicted and observed yields in the signal regions in the $N_{jet}= 3$ and $\geq 4$ seletions using the post-fit values of the nuisance parameters.

The measured inclusive ttgamma cross section in the fiducial phase space compared to the prediction from simulation using Madgraph_aMC@NLO at a center-of-mass energy of 13 TeV.

The measured inclusive ttgamma cross section in the fiducial phase space compared to the prediction from simulation using Madgraph_aMC@NLO at a center-of-mass energy of 13 TeV.

Summary of the measured cross section ratios with respect to the NLO cross section prediction for signal regions binned in the electron channel, muon channel and the combined single lepton measurement.

Summary of the measured cross section ratios with respect to the NLO cross section prediction for signal regions binned in the electron channel, muon channel and the combined single lepton measurement.

The unfolded differential cross sections for $p_{T}(\gamma)$ and the comparison to simulations.

The unfolded differential cross sections for $p_{T}(\gamma)$ and the comparison to simulations.

The unfolded differential cross sections for $|\eta(\gamma)|$ and the comparison to simulations.

The unfolded differential cross sections for $|\eta(\gamma)|$ and the comparison to simulations.

The unfolded differential cross sections for $\Delta R(l,\gamma)$ and the comparison to simulations.

The unfolded differential cross sections for $\Delta R(l,\gamma)$ and the comparison to simulations.

The covariance matrix of systematic uncertainties for the unfolded differential measurement for $p_{T}(\gamma)$.

The covariance matrix of systematic uncertainties for the unfolded differential measurement for $p_{T}(\gamma)$.

The covariance matrix of systematic uncertainties for the unfolded differential measurement for $|\eta(\gamma)|$.

The covariance matrix of systematic uncertainties for the unfolded differential measurement for $|\eta(\gamma)|$.

The covariance matrix of systematic uncertainties for the unfolded differential measurement for $\Delta R(l,\gamma)$.

The covariance matrix of systematic uncertainties for the unfolded differential measurement for $\Delta R(l,\gamma)$.

The covariance matrix of statistic uncertainties for the unfolded differential measurement for $p_{T}(\gamma)$.

The covariance matrix of statistic uncertainties for the unfolded differential measurement for $p_{T}(\gamma)$.

The covariance matrix of statistic uncertainties for the unfolded differential measurement for $|\eta(\gamma)|$.

The covariance matrix of statistic uncertainties for the unfolded differential measurement for $|\eta(\gamma)|$.

The covariance matrix of statistic uncertainties for the unfolded differential measurement for $\Delta R(l,\gamma)$.

The covariance matrix of statistic uncertainties for the unfolded differential measurement for $\Delta R(l,\gamma)$.

The correlation matrix of statistical uncertainties for the unfolded differential measurement for $p_{T}(\gamma)$.

The correlation matrix of statistical uncertainties for the unfolded differential measurement for $p_{T}(\gamma)$.

The correlation matrix of statistical uncertainties for the unfolded differential measurement for $|\eta(\gamma)|$.

The correlation matrix of statistical uncertainties for the unfolded differential measurement for $|\eta(\gamma)|$.

The correlation matrix of statistical uncertainties for the unfolded differential measurement for $\Delta R(l,\gamma)$.

The correlation matrix of statistical uncertainties for the unfolded differential measurement for $\Delta R(l,\gamma)$.

The correlation matrix of systematic uncertainties for the unfolded differential measurement for $p_{T}(\gamma)$.

The correlation matrix of systematic uncertainties for the unfolded differential measurement for $p_{T}(\gamma)$.

The correlation matrix of systematic uncertainties for the unfolded differential measurement for $|\eta(\gamma)|$.

The correlation matrix of systematic uncertainties for the unfolded differential measurement for $|\eta(\gamma)|$.

The correlation matrix of systematic uncertainties for the unfolded differential measurement for $\Delta R(l,\gamma)$.

The correlation matrix of systematic uncertainties for the unfolded differential measurement for $\Delta R(l,\gamma)$.

Summary of the one-dimensional intervals at 68 and 95% CL.

Summary of the one-dimensional intervals at 68 and 95% CL.

The observed and predicted post-fit yields for the combined Run 2 data set in the SR3 signal region for the electron channel.

The observed and predicted post-fit yields for the combined Run 2 data set in the SR3 signal region for the electron channel.

The observed and predicted post-fit yields for the combined Run 2 data set in the SR3 signal region for the muon channel.

The observed and predicted post-fit yields for the combined Run 2 data set in the SR3 signal region for the muon channel.

The observed and predicted post-fit yields for the combined Run 2 data set in the SR4p signal region for the electron channel.

The observed and predicted post-fit yields for the combined Run 2 data set in the SR4p signal region for the electron channel.

The observed and predicted post-fit yields for the combined Run 2 data set in the SR4p signal region for the muon channel.

The observed and predicted post-fit yields for the combined Run 2 data set in the SR4p signal region for the muon channel.

Negative log-likelihood ratio values with respect to the best fit value of the one-dimensional profiled scan for the Wilson coefficient $c_{tZ}$.

Negative log-likelihood ratio values with respect to the best fit value of the one-dimensional profiled scan for the Wilson coefficient $c_{tZ}$.

Negative log-likelihood ratio values with respect to the best fit value of the one-dimensional profiled scan for the Wilson coefficient $c^{I}_{tZ}$.

Negative log-likelihood ratio values with respect to the best fit value of the one-dimensional profiled scan for the Wilson coefficient $c^{I}_{tZ}$.

Negative log-likelihood ratio values with respect to the best fit value of the one-dimensional scan for the Wilson coefficient $c_{tZ}$.

Negative log-likelihood ratio values with respect to the best fit value of the one-dimensional scan for the Wilson coefficient $c_{tZ}$.

Negative log-likelihood ratio values with respect to the best fit value of the one-dimensional scan for the Wilson coefficient $c^{I}_{tZ}$.

Negative log-likelihood ratio values with respect to the best fit value of the one-dimensional scan for the Wilson coefficient $c^{I}_{tZ}$.

Negative log-likelihood ratio values with respect to the best fit value of the two-dimensional scan for the Wilson coefficients $c_{tZ}$ and $c^{I}_{tZ}$.

Negative log-likelihood ratio values with respect to the best fit value of the two-dimensional scan for the Wilson coefficients $c_{tZ}$ and $c^{I}_{tZ}$.