The measured $E_{T}^{miss}$ distribution in the Z/$\gamma ^{*}$($\rightarrow \mu \mu$)+jets control region, for the $E_{T}^{miss}$ > 250GeV inclusive selection, compared to the background predictions. The latter include the global normalization factors extracted from the fit. The last bin of the distribution contains overflows.

The measured leading jet $p_{T}$ distribution in the Z/$\gamma ^{*}$($\rightarrow \mu \mu$)+jets control region, for the $E_{T}^{miss}$ > 250GeV inclusive selection, compared to the background predictions. The latter include the global normalization factors extracted from the fit. The last bin of the distribution contains overflows.

The measured $E_{T}^{miss}$ distribution in the top control region, for the $E_{T}^{miss}$ > 250GeV inclusive selection, compared to the background predictions. The latter include the global normalization factors extracted from the fit. The last bin of the distribution contains overflows.

The measured leading jet $p_{T}$ distribution in the top control region, for the $E_{T}^{miss}$ > 250GeV inclusive selection, compared to the background predictions. The latter include the global normalization factors extracted from the fit. The last bin of the distribution contains overflows.

Measured distribution of the $E_{T}^{miss}$ for the $E_{T}^{miss}$ > 250GeV selection compared to the SM predictions. The latter are normalized with normalization factors as determined by the global fit that considers exclusive $E_{T}^{miss}$ regions. The last bin of the distribution contains overflows.

Measured distribution of the leading jet $p_{T}$ for the $E_{T}^{miss}$ > 250GeV selection compared to the SM predictions. The latter are normalized with normalization factors as determined by the global fit that considers exclusive $E_{T}^{miss}$ regions. The last bin of the distribution contains overflows.

Measured distribution of the leading jet $|\eta|$ for the $E_{T}^{miss}$ > 250GeV selection compared to the SM predictions. The latter are normalized with normalization factors as determined by the global fit that considers exclusive $E_{T}^{miss}$ regions. The last bin of the distribution contains overflows.

Measured distribution of the jet multiplicity for the $E_{T}^{miss}$ > 250GeV selection compared to the SM predictions. The latter are normalized with normalization factors as determined by the global fit that considers exclusive $E_{T}^{miss}$ regions. The last bin of the distribution contains overflows.

The expected $95\%$ CL exclusion limit for a simplified model of dark matter production involving an axial-vector operator, Dirac DM and couplings $g_{q} = 0.25$ and $g_{\chi} = 1$ as a function of the assumed mediator mass m$_{Z_{A}}$ and the dark matter mass m$_{\chi}$.

The measured $E_{T}^{miss}$ distribution in the W($\rightarrow \mu \nu$)+jets control region, for the $E_{T}^{miss}$ > 250GeV inclusive selection, compared to the background predictions. The latter include the global normalization factors extracted from the fit. The last bin of the distribution contains overflows.

The observed $95\%$ CL exclusion limit for a simplified model of dark matter production involving an axial-vector operator, Dirac DM and couplings $g_{q} = 0.25$ and $g_{\chi} = 1$ as a function of the assumed mediator mass m$_{Z_{A}}$ and the dark matter mass m$_{\chi}$.

The observed $90\%$ CL exclusion limit on the spin-dependent WIMP–proton scattering cross section in the context of the simplified model with axial-vector couplings, assuming minimal mediator width and the coupling values $g_{q} = 0.25$ and $g_{\chi} = 1$.

The expected $95\%$ CL exclusion limit for a simplified model of dark matter production involving a vector operator, Dirac DM and couplings $g_{q} = 0.25$ and $g_{\chi} = 1$ as a function of the assumed mediator mass m$_{Z_{V}}$ and the dark matter mass m$_{\chi}$.

The observed $95\%$ CL exclusion limit for a simplified model of dark matter production involving a vector operator, Dirac DM and couplings $g_{q} = 0.25$ and $g_{\chi} = 1$ as a function of the assumed mediator mass m$_{Z_{V}}$ and the dark matter mass m$_{\chi}$.

The expected and observed $95\%$ CL limits on the signal strength $\mu = \sigma^{95\% CL}/\sigma$ as a function of the mediator mass for a very light WIMP, in a model with spin-0 pseudoscalar mediator and $g_{q}=g_{\chi}=1.0$.

The expected and observed $95\%$ CL limits on the signal strength $\mu = \sigma^{95\% CL}/\sigma$ as a function of the WIMP mass for $m_{Z_{P}}=10$ GeV, in a model with spin-0 pseudoscalar mediator and $g_{q}=g_{\chi}=1.0$.

The expected exclusion contour at $95\%$ CL in the m$_{\eta}$–m$_{\chi}$ parameter plane for the coloured scalar mediator model, with minimal width and coupling set to $g=1$.

The observed exclusion contour at $95\%$ CL in the m$_{\eta}$–m$_{\chi}$ parameter plane for the coloured scalar mediator model, with minimal width and coupling set to $g=1$.

The expected excluded region at the $95\%$ CL in the ($\tilde{t}_{1}$,$\chi^{0}_{1}$) mass plane for the decay channel $\tilde{t}_{1} \rightarrow c + \chi^{0}_{1}$ (B = $100\%$).

The observed excluded region at the $95\%$ CL in the ($\tilde{t}_{1}$,$\chi^{0}_{1}$) mass plane for the decay channel $\tilde{t}_{1} \rightarrow c + \chi^{0}_{1}$ (B = $100\%$).

The expected excluded region at the $95\%$ CL in the ($\tilde{t}_{1}$,$\chi^{0}_{1}$) mass plane for the decay channel $\tilde{t}_{1} \rightarrow b + ff' + \chi^{0}_{1}$ (B = $100\%$).

The observed excluded region at the $95\%$ CL in the ($\tilde{t}_{1}$,$\chi^{0}_{1}$) mass plane for the decay channel $\tilde{t}_{1} \rightarrow b + ff' + \chi^{0}_{1}$ (B = $100\%$).

The expected exclusion plane at $95\%$ CL as a function of sbottom and neutralino masses for the decay channel $\tilde{b}_{1} \rightarrow b + \chi^{0}_{1}$ (B = $100\%$).

The observed exclusion plane at $95\%$ CL as a function of sbottom and neutralino masses for the decay channel $\tilde{b}_{1} \rightarrow b + \chi^{0}_{1}$ (B = $100\%$).

The expected exclusion region at $95\%$ CL as a function of squark mass and the squark-neutralino mass difference for $\tilde{q}_{1} → q + \chi^{0}_{1}$ (q =u,d,c,s).

The observed exclusion region at $95\%$ CL as a function of squark mass and the squark-neutralino mass difference for $\tilde{q}_{1} → q + \chi^{0}_{1}$ (q =u,d,c,s).

Expected and observed $95\%$ CL lower limits on the fundamental Planck scale in 4+n dimensions, M$_D$, as a function of the number of extra dimensions.

Expected and observed $95\%$ CL upper limit on the signal strength $\mu$ in the hypothesis of an axial-vector mediator, g$_{q}=0.25$, g$_{\chi}=1.0$ and minimal mediator width, as a function of the assumed mediator and DM masses.

Observed $90\%$ CL exclusion limit on the spin-dependent WIMP–neutron scattering cross section in the context of the simplified model with axial-vector couplings, assuming minimal mediator width and the coupling values $g_{q}=0.25$ and $g_{\chi}=1$.

Expected and observed $95\%$ CL upper limit on the signal strength $\mu$ in the hypothesis of a pseudoscalar mediator, $g_{q}=g_{\chi}=1.0$ and minimal mediator width, as a function of the assumed mediator and DM masses.

Comparison of data with estimated backgrounds in the $\tilde{E}_\text{T}^\text{miss}$ distribution with the TZCR1 event selection except for the requirement on $\tilde{E}_\text{T}^\text{miss}$. The variables $\tilde{E}_\text{T}^\text{miss}$ and $\tilde{m}_\text{T}$ are constructed in the same way as $E_\text{T}^\text{miss}$ and $m_\text{T}$ but treating the leading photon transverse momentum as invisible. The predicted backgrounds are scaled with normalization factors. The uncertainty band includes statistical and all experimental systematic uncertainties. The last bin includes overflow.

Comparison of data with estimated backgrounds in the $\tilde{m}_\text{T}$ distribution with the TZCR1 event selection except for the requirement on $\tilde{m}_\text{T}$. The variables $\tilde{E}_\text{T}^\text{miss}$ and $\tilde{m}_\text{T}$ are constructed in the same way as $E_\text{T}^\text{miss}$ and $m_\text{T}$ but treating the leading photon transverse momentum as invisible. The predicted backgrounds are scaled with normalization factors. The uncertainty band includes statistical and all experimental systematic uncertainties. The last bin includes overflow.

Comparison of the observed data ($n_\text{obs}$) with the predicted background ($n_\text{exp}$) in the validation and signal regions. The background predictions are obtained using the background-only fit configuration. The bottom panel shows the significance of the difference between data and predicted background, where the significance is based on the total uncertainty ($\sigma_\text{tot}$).

Jet multiplicity distributions for events where exactly two signal leptons are selected. No correction factors are included in the background normalizations. The uncertainty band includes statistical and all experimental systematic uncertainties. The last bin includes overflow.

Jet multiplicity distributions for events where exactly one lepton plus one $\tau$ candidate are selected. No correction factors are included in the background normalizations. The uncertainty band includes statistical and all experimental systematic uncertainties. The last bin includes overflow.

The $E_\text{T}^\text{miss}$ distribution in SR1. In the plot, the full event selection in the corresponding signal region is applied, except for the requirement on $E_\text{T}^\text{miss}$. The predicted backgrounds are scaled with normalization factors. The uncertainty band includes statistical and all experimental systematic uncertainties. The last bin contains the overflow. Benchmark signal models are overlaid for comparison. The benchmark models are specified by the gluino and stop masses, given in TeV in the table.

The $m_\text{T}$ distribution in SR1. In the plot, the full event selection in the corresponding signal region is applied, except for the requirement on $m_\text{T}$. The predicted backgrounds are scaled with normalization factors. The uncertainty band includes statistical and all experimental systematic uncertainties. The last bin contains the overflow. Benchmark signal models are overlaid for comparison. The benchmark models are specified by the gluino and stop masses, given in TeV in the table.

Expected (black dashed) 95% excluded regions in the plane of $m_{\tilde{g}}$ versus $m_{\tilde{t}_1}$ for gluino-mediated stop production.

Observed (red solid) 95% excluded regions in the plane of $m_{\tilde{g}}$ versus $m_{\tilde{t}_1}$ for gluino-mediated stop production.

Expected (black dashed) 95% excluded regions in the plane of $m_{\tilde{t}_1}$ versus $m_{\tilde{\chi}_1^0}$ for direct stop production.

Observed (red solid) 95% excluded regions in the plane of $m_{\tilde{t}_1}$ versus $m_{\tilde{\chi}_1^0}$ for direct stop production.

The expected upper limits on $T$ quark pair production times the squared branching ratio for $T \rightarrow tZ$ as a function of the $T$ quark mass.

The observed upper limits on $T$ quark pair production times the squared branching ratio for $T \rightarrow tZ$ as a function of the $T$ quark mass.

The expected limits on $T$ quarks as a function of the branching ratios $B\left(T \rightarrow bW\right)$ and $B\left(T \rightarrow tH\right)$ for a $T$ quark with a mass of 800 GeV. The $T$ is assumed to decay in three possible ways: $T \to tZ$, $T \to tH$, and $T \to bW$.

The observed limits on $T$ quarks as a function of the branching ratios $B\left(T \rightarrow bW\right)$ and $B\left(T \rightarrow tH\right)$ for a $T$ quark with a mass of 800 GeV. The $T$ is assumed to decay in three possible ways: $T \to tZ$, $T \to tH$, and $T \to bW$.

The $m_\text{T}$ distribution in the WVR2-tail validation region which has the same preselection and jet $p_\text{T}$ requirements as SR2.

The $am_\text{T2}$ distribution in the WVR2-tail validation region which has the same preselection and jet $p_\text{T}$ requirements as SR2.

Large-radius jet mass ($R=1.2$), decomposed into the number of small-radius jet constituents. The lower panel shows the ratio of the total data to the total prediction (summed over all jet multiplicities). Events are required to have one lepton, four jets with $p_\text{T}>80,50,40,40$ GeV, at least one $b$-tagged jet, $E_\text{T}^\text{miss}>200$ GeV, and $m_\text{T}>30$ GeV.

Distribution of $m_\text{T2}^\tau$ in data for a selection enriched in $t\bar{t}$ events with one hadronically decaying $\tau$. Events that have no hadronic $\tau$ candidate (that passes the Loose identification criteria, as well as other requirements) are not shown in the plot.

Upper limits on the model cross-section in units of pb for the gluino-mediated stop models.

Upper limits on the model cross-section in units of pb for the models with direct stop pair production.

Illustration of the best expected signal region per signal grid point for the gluino-mediated stop models. This mapping is used for the final combined exclusion limits.

Illustration of the best expected signal region per signal grid point for models with direct stop pair production. This mapping is used for the final combined exclusion limits.

Expected $CL_s$ values for the gluino-mediated stop models.

Observed $CL_s$ values for the gluino-mediated stop models.

Expected $CL_s$ values for the direct stop pair production models.

Observed $CL_s$ values for the direct stop pair production models.

Expected limit using SR1 for models with direct stop pair production and an unpolarized stop (and bino LSP).

Expected limit using SR1 for models with direct stop pair production with $\tilde{t}_1=\tilde{t}_L$ (and bino LSP).

Expected limit using SR1 for models with direct stop pair production with $\tilde{t}_1\sim\tilde{t}_R$ (and bino LSP).

Observed limit using SR1 for models with direct stop pair production and an unpolarized stop (and bino LSP).

Observed limit using SR1 for models with direct stop pair production with $\tilde{t}_1=\tilde{t}_L$ (and bino LSP).

Observed limit using SR1 for models with direct stop pair production with $\tilde{t}_1\sim\tilde{t}_R$ (and bino LSP).

Expected limit using SR2 for models with direct stop pair production and an unpolarized stop (and bino LSP).

Expected limit using SR2 for models with direct stop pair production with $\tilde{t}_1=\tilde{t}_L$ (and bino LSP).

Expected limit using SR2 for models with direct stop pair production with $\tilde{t}_1\sim\tilde{t}_R$ (and bino LSP).

Observed limit using SR2 for models with direct stop pair production and an unpolarized stop (and bino LSP).

Observed limit using SR2 for models with direct stop pair production with $\tilde{t}_1=\tilde{t}_L$ (and bino LSP).

Observed limit using SR2 for models with direct stop pair production with $\tilde{t}_1\sim\tilde{t}_R$ (and bino LSP).

Expected limit using SR1+SR2 (best expected) for models with direct stop pair production and an unpolarized stop (and bino LSP).

Expected limit using SR1+SR2 (best expected) for models with direct stop pair production with $\tilde{t}_1=\tilde{t}_L$ (and bino LSP).

Expected limit using SR1+SR2 (best expected) for models with direct stop pair production with $\tilde{t}_1\sim\tilde{t}_R$ (and bino LSP).

Observed limit using SR1+SR2 (best expected) for models with direct stop pair production and an unpolarized stop (and bino LSP).

Observed limit using SR1+SR2 (best expected) for models with direct stop pair production with $\tilde{t}_1=\tilde{t}_L$ (and bino LSP).

Observed limit using SR1+SR2 (best expected) for models with direct stop pair production with $\tilde{t}_1\sim\tilde{t}_R$ (and bino LSP).

Acceptance for SR1 in the gluino-mediated stop models. The acceptance is defined as the fraction of signal events that pass the analysis selection performed on generator-level objects, therefore emulating an ideal detector with perfect particle identification and no measurement resolution effects.

Acceptance for SR1 in the direct stop pair production. The acceptance is defined as the fraction of signal events that pass the analysis selection performed on generator-level objects, therefore emulating an ideal detector with perfect particle identification and no measurement resolution effects.

Acceptance for SR2 in the gluino-mediated stop models. The acceptance is defined as the fraction of signal events that pass the analysis selection performed on generator-level objects, therefore emulating an ideal detector with perfect particle identification and no measurement resolution effects.

Acceptance for SR2 in the direct stop pair production. The acceptance is defined as the fraction of signal events that pass the analysis selection performed on generator-level objects, therefore emulating an ideal detector with perfect particle identification and no measurement resolution effects.

Acceptance for SR3 in the gluino-mediated stop models. The acceptance is defined as the fraction of signal events that pass the analysis selection performed on generator-level objects, therefore emulating an ideal detector with perfect particle identification and no measurement resolution effects.

Acceptance for SR3 in the direct stop pair production. The acceptance is defined as the fraction of signal events that pass the analysis selection performed on generator-level objects, therefore emulating an ideal detector with perfect particle identification and no measurement resolution effects.

Efficiency for SR1 in the gluino-mediated stop models. The efficiency is the ratio between the expected signal rate calculated with simulated data passing all the reconstruction level cuts applied to reconstructed objects, and the signal rate for an ideal detector (with perfect particle identification and no measurement resolution effects).

Efficiency for SR1 in the direct stop pair production. The efficiency is the ratio between the expected signal rate calculated with simulated data passing all the reconstruction level cuts applied to reconstructed objects, and the signal rate for an ideal detector (with perfect particle identification and no measurement resolution effects).

Efficiency for SR2 in the gluino-mediated stop models. The efficiency is the ratio between the expected signal rate calculated with simulated data passing all the reconstruction level cuts applied to reconstructed objects, and the signal rate for an ideal detector (with perfect particle identification and no measurement resolution effects).

Efficiency for SR2 in the direct stop pair production. The efficiency is the ratio between the expected signal rate calculated with simulated data passing all the reconstruction level cuts applied to reconstructed objects, and the signal rate for an ideal detector (with perfect particle identification and no measurement resolution effects).

Efficiency for SR3 in the gluino-mediated stop models. The efficiency is the ratio between the expected signal rate calculated with simulated data passing all the reconstruction level cuts applied to reconstructed objects, and the signal rate for an ideal detector (with perfect particle identification and no measurement resolution effects).

Efficiency for SR3 in the direct stop pair production. The efficiency is the ratio between the expected signal rate calculated with simulated data passing all the reconstruction level cuts applied to reconstructed objects, and the signal rate for an ideal detector (with perfect particle identification and no measurement resolution effects).