> home > Feature Articles
 
Recent Updates on the Standard Model Higgs Boson Measurements from the ATLAS and CMS Experiments
Song-Ming Wang
File 1 : Vol27_No6_Feature Articles-3.pdf (0 byte)

Recent Updates on the Standard Model Higgs Boson Measurements
from the ATLAS and CMS Experiments

SONG-MING WANG
INSTITUTE OF PHYSICS, ACADEMIA SINICA

ABSTRACT

This report presents the latest results from the ATLAS and CMS experiments on the measurements of the Standard Model Higgs boson by using the proton-proton collisions produced by the Large Hadron Collider during the first two years of Run 2 data taking.

INTRODUCTION

A Standard Model (SM)-like Higgs boson particle was discovered by the ATLAS [1] and CMS [2] experiments in 2012 in the proton-proton (p-p) collision events produced by the Large Hadron Collider (LHC) [3] at √s=7,8 TeV during the Run 1 data taking operation [4,5]. The integrated luminosities of the Run 1 data samples, collected individually by each experiment, were about 25 fb-1. The mass of this new particle was measured to be mH=125.09 ± 0.24 GeV [6]. Many measurements that came after the discovery further strengthened the hypothesis that the new particle was the SM Higgs boson.

The SM Higgs boson can be produced via several production channels in p-p collisions, and it can undergo various decay modes (e.g., into pairs of fermions or gauge bosons). For a Higgs mass of mH=125 GeV, the gluon fusion (ggF) process is predicted to have the highest production rate at the LHC, followed by vector boson fusion (VBF), the WH and ZH associated production (VH), and then the ttH associated production. There are other Higgs boson production processes (e.g., production in association with a b-quark or with a single top quark). However, their production rates are much smaller. At mH=125 GeV, the Higgs boson is predicted to decay predominantly into a pair of b-quarks (BR(Hbb̄)~58%). The predicted decay rates to ττ and μμ are, respectively, about ~6% and ~0.02%, and into WW, ZZ and γγ at ~20%, ~3% and ~0.2%, respectively.

The discovery of this new particle in Run 1 was through the channels where the Higgs boson is produced mainly through the ggF and VBF processes, and decays into a pair of gauge bosons (Hγγ, WW, ZZ). The Hτ+τ- decay mode was observed in Run 1 when combining the search results from ATLAS and CMS [7]. The other predicted production and decay channels had not yet been observed. Thus it was essential to study whether this SM like Higgs particle could be produced through these other processes, and whether it would also decay directly into other fermionic channels. Any deviation from these predictions could indicate the presence of new physics.

The center-of-mass energy at the LHC was increased to 13 TeV for Run 2 data taking and the predicted SM Higgs production rates, depending on the production process, increased by factors of ~2-4 compared to Run 1. Both ATLAS and CMS detectors recorded about 36 fb-1 of data during 2015 and 2016 data taking. Researchers have analyzed this larger data sample to perform more systematic measurements of the properties of this new particle, and to search for it in the channels that have not yet been observed. Their latest results with the Run 2 data are summarized in this article. Many of the results presented are on the limits of the production cross section times branching ratio (σ×BR) at 95% confidence level (CL), and the measured signal strength μ, which is defined as the ratio of the measured signal rate with respect to the SM predicted rate.

H γγ, ZZ

The ATLAS and CMS experiments have re-visited the discovery channels by measuring the Higgs boson production in the Hγγ and HZZ decay channels with Run 2 data samples (L~36 fb-1) to perform more precise measurements of the Higgs boson production rates, its mass, and its properties as a function of several kinematic quantities.

By considering all the selected events, the Higgs signal is extracted from the di-photon mass (mγγ) distribution for Hγγ, and from the four-lepton invariant mass (m4l) distribution for HZZ→4l. These distributions are shown in Fig. 1. For HZZ→4l, the measured signal strengths from ATLAS and CMS are μ=1.28+0.18-0.17(stat.)+0.08-0.06(exp.)+0.08-0.06(th.) and μ=1.05+0.15-0.14(stat.)+0.11-0.09(syst.), respectively [8,9]. In the case of Hγγ, CMS measures a signal strength of μ=1.16+0.11-0.10(stat.)+0.09-0.08(syst.)+0.06-0.05(th.), and ATLAS measures μ=0.99+0.12-0.11(stat.)+0.06-0.05(exp.)+0.06-0.05(th.) [10,11]. The signal strength parameters are measured for mH=125.09 GeV. Despite having lower signal purity for Hγγ, its signal strength's precisions are similar to the ones from HZZ→4l. The uncertainties of the measured signal strengths from Run 2 of both experiments are reduced by half when compared to the Run 1 measurements [12-15].



Fig. 1: (Left) Di-photon invariant mass mγγ distribution for Higgs measurements in the H→γγ decay channel. (Right) Four-lepton invariant mass m4l distribution for Higgs measurement in the H→ZZ→4l decay channel.

The inclusive signal yield in each decay channel is also used to measure the total cross section of Higgs production (ppH+X) at the LHC. The measurements from ATLAS for √s=13 TeV are shown in Fig. 2 for Hγγ, HZZ→4l and combined [16]. The figure also shows the past measurements from Run 1 at √s=7 and 8 TeV, and the SM prediction. The measurements at the three center-of-mass energies are compatible with the SM prediction.





Fig. 2: Measurements of the total cross section of Higgs production (ppH+X) at the LHC from ATLAS.

The Higgs boson mass was measured in Run 1 by ATLAS and CMS, and the combined measurement is mH=125.09 ± 0.24 GeV. This measurement has been updated by both experiments for Run 2 by using the Hγγ and HZZ→4l decay channels. ATLAS' measurements, which are shown in Fig. 3, are derived from a combined fit to the invariant mass spectra of both decay channels. The measured mass from HZZ→4l (Hγγ) is mH=124.88 ± 0.37(stat.) ± 0.05(syst.) GeV (mH=125.11 ± 0.21(stat.) ± 0.36(syst.) GeV). The combined mass is mH=124.98 ± 0.19(stat.) ± 0.21(syst.) GeV [17]. The measurements between the two decay channels are compatible to 0.4σ, and the uncertainty of ATLAS' Run 2 combined measurement is similar to the Run 1 ATLAS and CMS combined measurement. CMS has also updated the Higgs mass measurement in the HZZ→4l channel. A multi-dimensional fit to the four-lepton invariant mass m4l, mass uncertainty and ZZ background discriminator has been performed. The measured mass is mH=125.26 ± 0.20(stat.) ± 0.08(syst.) GeV [11]. This single measurement is competitive with the ATLAS+CMS Run 1 measurement.





Fig. 3: Measurements of Higgs boson mass by ATLAS in the H→γγ and H→ZZ→4l decay channels.

The measurements in these two decay channels have been also used by both experiments to extract the signal strengths for various Higgs production processes. For these measurements, the selected events are grouped into several exclusive categories where each category targets different production mechanism. The measured signal strengths of several production processes, by using the Hγγ decay channel by CMS, are shown in Fig. 4 [9]. ATLAS' measurements, combining both decay channels, are also shown [16]. For both experiments, the measured signal strength for ggF Higgs production is very consistent with SM. The signal strength of the VBF process, measured by ATLAS, is higher than the SM prediction by ~2σ.





Fig. 4: Signal strength values for various Higgs production processes measured in H→γγ and H→ZZ→4l decay channels.

The larger Run 2 data samples allow ATLAS and CMS to probe more kinematic properties of the Higgs boson by measuring the differential cross section of several kinematic quantities. The differential cross section distribution as a function of the transverse momentum of the reconstructed Higgs candidates (for HZZ→4l) is shown in Fig. 5 (Left) [11]. This measurement is sensitive to perturbative QCD modeling of the ggF process, and to new heavy particles coupling to the Higgs boson. The measurement of differential cross section distribution as a function of |Δyγγ| (rapidity difference between the leading and sub-leading photons) for Hγγ [8], shown in Fig. 5 (Right), is sensitive to the spin of the Higgs boson. All the measurements show well compatibility with the SM.

H μ+μ-

The Hμ+μ- decay mode has not been observed at the LHC. Its search allows one to probe the Higgs boson coupling to second generation fermions. ATLAS had conducted such a search in Run 1 [18] and has extended the search with a larger data sample of Run 2 with an integrated luminosity of 36.1 fb-1 [19]. Since the final state, which consists of a pair of oppositely charged muons, is relatively clean of background, ATLAS chose to perform the search in the two leading production channels (ggF and VBF). This helps to make up for the small Hμ+μ- decay rate.



Fig. 5: (Left) Differential cross section distribution as a function of the transverse momentum of the reconstructed Higgs candidates for H→ZZ→4l analysis from CMS. (Right) Differential cross section distribution as a function of |Δyγγ| for H→γγ analysis from ATLAS.

The analysis selects events with exactly two opposite charged muons and classifies them into categories that are sensitive to signal events produced in VBF and ggF processes. The signal event selection priority is first given to the VBF production by requiring two high transverse momentum jets in opposite detector hemispheres. These events then undergo a multivariate analysis (MVA) where a boosted decision tree (BDT) algorithm is trained on several kinematic variables to enrich the VBF signal in the high region of the BDT output distribution, which is defined to be the VBF signal region and is divided into "loose" and "tight" categories. The events that fall outside the VBF signal region in the BDT distribution, or fail the two-jet requirement, are then optimized for the search in the ggF channel. These events are grouped into different categories based on the transverse momentum (pTμμ) and pseudo rapidity (ημμ) of the μ+μ- system. The signal is searched for in the di-muon invariant mass (mμμ) distributions of the selected events by performing simultaneous fits to the mμμ distributions from all the VBF and ggF categories. The mμμ distribution of the "tight" VBF category is shown in Fig. 6. The extracted signal strength value is μ = -0.1 ± 1.5 and the observed (expected) upper limit at 95% confidence level on the production cross section times branching fraction is 3.0 (3.1) times the SM prediction. When combined with Run 1 search results, the combined signal strength value is μ = -0.1 ± 1.4 and the observed (expected) upper limit on σ×BR is 2.8 (2.9) times the SM prediction. The signal strength and cross section limits are evaluated at mH=125 GeV. The sensitivity of this search is currently limited by the data's statistical uncertainty.





Fig. 6: mμμ distribution of the "tight" VBF category in the H→μ+μ- decay channel.

H τ+τ-

ATLAS and CMS searched for the Higgs boson in the di-tau decay channel in Run 1. Both experiments observed evidence of such decay in their data. ATLAS had an observed significance of 4.5 standard deviations [20] and CMS reported an observed excess of 3.2 standard deviations [21]. Their Run 1 combined results yielded an observed significance of 5.5 standard deviations [7].

This search has been repeated by CMS by analyzing its Run 2 data sample with an integrated luminosity of 35.9 fb-1 [22]. The analysis is optimized for Higgs bosons produced through the VBF and ggF process. The leptonic (l = e, μ) and hadronic (h) tau decay modes are considered in the search and the di-tau ττ final states include τeτμ, τeτh, τμτh and τhτh. The τeτe, τμτμ are excluded due to the large background source from Z→e+e-, μ+μ- production.

The di-tau invariant mass mττ is a powerful discriminating variable to search for the Higgs boson in this decay channel. However, the neutrinos from tau lepton decays carry a large fraction of the tau lepton energy and reduce the discriminating power. An algorithm SVFIT [23], which combines the missing transverse energy with the four-vectors of both selected tau candidates, is used to calculate a more accurate estimate of mττ mass of the Higgs boson.

The selected events are grouped into three categories. The first category "0-jet", which consists of events with no jet with transverse momentum pT>30 GeV, targets the Higgs boson events produced via ggF. The second category "VBF" targets Higgs boson events produced via VBF by requiring selected events to have at least two jets with pT>30 GeV. The last category "boosted" consists of the remaining events that are not selected in the other two categories. The signal events in this category are still dominated by Higgs events produced via ggF.

The main background sources are from Z(ττ)+jets production, W+jets production, tt̄ production and background from jets faking taus.

The selected events in each category are sub-divided into groups, with varying signal sensitivities, based on different regions of a kinematic variable. The choice of the kinematic variable depends on the category and the di-tau decay channel. The Higgs signal is then extracted from the mττ distribution in each of these groups. For example, in the VBF category and for τhτh decay channel, the signal is searched for in the mττ distributions from different bins of mJJ (mass of the two leading jets in the VBF category), as shown in Fig. 7.





Fig. 7: mττ distributions in various bins of mJJ for the selected events in the VBF category and for the τhτh decay channel.

By grouping the events in the signal regions based on the quantity log10(S/(S+B)), where S and B, respectively, are the signal and background yields in each bin of the mττ distributions, one obtains a distribution (Fig. 8) where it shows an excess of observed data events over the predicted SM background. The significance of the observed excess is 4.9 standard deviations, with an expected significance of 4.7 standard deviations. The extracted signal strength is μ=1.09+0.27-0.26 at mH=125.09 GeV. By combining these findings with CMS Run 1 results, the observed and expected significance is 5.9 standard deviations. The combined signal strength is μ=0.98±0.18 at mH=125.09 GeV. This is the first Hτ+τ- observation by a single experiment!



Fig. 8: Distribution of log10(S/(S+B)), where S and B, respectively, are the signal and background yields in each bin of the mττ distributions in all signal regions.

INCLUSIVE BOOSTED H bb̄

During LHC Run 1 data taking, the Higgs boson was discovered by ATLAS and CMS in the bosonic decay channels. The Hτ+τ- channel was discovered when combining both experiments' Run 1 results, and was recently observed within CMS when combining its Run 1 and Run 2 results (as described in the previous section). However, the Hbb̄ decay channel has not yet been observed despite its large decay branching ratio. The VH associated production is the most sensitive method to search for this Higgs decay. Other production channels have also been searched for by both LHC experiments.

Recently, CMS performed an inclusive search for Hbb̄ at the high transverse momentum region of the Higgs boson [24]. In this case, the Higgs system is highly boosted such that the two b-jets from the Higgs decay are closed to each other and are reconstructed as a single large-R jet. The signal contribution is mostly from ggF production. The main experimental challenges are the large multi-jet background and the trigger requirements needed to reduce the data taking rate.

The signal events are selected by requiring a large-R jet with transverse momentum pT>450 GeV and pass a double b-tagged discriminator threshold. Events with the presence of charged leptons or with high missing transverse energy are vetoed to suppress background contributions from SM electroweak processes and top quark production. The signal is searched for in the large-R jet mass (mSD) distribution (shown in Fig. 9). A Zbb̄ mass peak is observed in the mSD distribution with an observed (expected) significance of 5.1 (5.8) standard deviations. The signal strength is μZ = 0.78 ± 0.14(stat.) +0.19-0.13(syst.). The measured Higgs boson signal strength is μH=2.3 ± 1.5(stat.) +1.0-0.4(syst.) at mH=125 GeV. The observed (expected) significance is 1.5 (0.7) standard deviations. This is the first analysis at the LHC to search for Hbb̄ in the inclusive channel, and this analysis produced the first observation of Zbb̄ in the boosted channel.





Fig. 9: Large-R jet mass (mSD) distribution for the Higgs boson search in the inclusive boosted Hbb̄ channel.

H bb̄ IN VBF PLUS PHOTON PRODUCTION

The Higgs boson in the Hbb̄ decay mode has also been searched for by the ATLAS and CMS experiments in the VBF production channel in Run 1 [25,26]. In Run 2, ATLAS updated this search with an inclusion of a photon in the final state. The search is performed on a data sample with an integrated luminosity of 12.6 fb-1 [27]. The photon may be radiated from an internal W boson or from an incoming or outgoing quark. The extra photon in the final state helps to reduce the dominant multi-jet background, and it also provides an extra handle to trigger on the signal.

The signal events are selected by requiring two identified b-jets, two forward jets and a central photon. The other major background source is from Z(bb)+jets production. A BDT is trained to separate the signal against the multi-jet background, and the BDT output distribution is divided into three regions (low, medium, high) of varying signal sensitivities. The signal is searched for in the di-b-jet invariant mass mbb distribution of each BDT region. Fig. 10 shows the mbb distribution in the high BDT region. The multi-jet background contribution is estimated by fitting the low and high side-band regions of the mbb distribution in the data, and the Z(bb)+jets background is estimated from simulations. In the absence of observing the Hbb̄ signal, the analysis set a preliminary observed (expected) 95% CL upper limit on the Higgs production cross section times branching ratio of 4.0 (6.0+2.3-1.7) times the SM expectation for mH=125 GeV.





Fig. 10: mbb distribution in the high BDT region of the Hbb̄ decay for the search in the VBF plus photon production channel.

H bb̄ IN WH AND ZH ASSOCIATED PRODUCTION

As mentioned in earlier section, the VH (V is W or Z) associated production is the most sensitive method to search for the Higgs boson in the Hbb̄ decay mode. ATLAS and CMS had performed searches in this production channel in Run 1 [28,29]. The observed (expected) significance from ATLAS is 1.7 (2.7) standard deviations, whereas CMS obtained 2.0 (2.5) standard deviations for the observed (expected) significance.

Both experiments have recently updated the search by analyzing ~36 fb-1 of Run 2 data sample collected in 2015 and 2016 [30,31]. The signal event selection and analysis strategies are quite similar between the two searches. The selected signal events require two identified b-jets and that the final state contains 0, 1 and 2 charged leptons (electrons or muons), which correspondingly target the decays Zvv, Wlv and Zll. The searches are conducted at medium and high transverse momentum of the vector boson (pTV>~150-170 GeV for 0-lepton, pTV>~100-150 GeV for 1-lepton and pTV>~50-75 GeV for 2-lepton) where the signal purity is higher. The selected events of each leptonic channel are divided into several categories to maximize the search sensitivity. The division is based on the transverse momentum of the vector boson. For ATLAS, the selected events are also categorized based on the number of reconstructed jets in the final state. Several control regions are also defined for the search by both experiments. These control regions are designed to help determine the normalization of the main background processes, and to validate the modeling of the distributions of variables most relevant to the analysis.

The searches from both experiments separate the signal from the background by constructing multivariate discriminants. A BDT algorithm is trained separately for each category of each leptonic channel on several kinematic variables that are tailored for that category. The BDT output distribution is used as the main discriminant.

The mbb mass distribution of the two b-tagged-jet system is one of the key discriminating variables. Both searches applied corrections to improve the mbb mass resolution and its mean value. CMS applies a multivariate regression technique [32], where a BDT is trained on b-jets from simulated tt events, to correct the measured energy of each individual b-tagged jet. This correction improves the mbb resolution by ~15% (shown in Fig. 11, Left). For ATLAS, several corrections are applied to the b-tagged jet's measured energy. These corrections include "muon-in-jet", "b-jet energy respond correction" and "kinematic likelihood fit". Details of these corrections are described in [30]. These corrections improve the mbb resolution by ~18-40% (shown in Fig. 11, Right).





Fig. 11: mbb distribution with additional corrections applied to improve the mass resolution. (Left) CMS, (Rigth) ATLAS.

The BDT output distributions from a few categories of the signal region, from both searches, are shown in Fig. 12. Fig. 12 Left and Right, respectively, are the BDT output distributions of the 0-lepton and 1-lepton channel in the 2-jet final state from ATLAS. Fig. 12 Right is the BDT output distribution of the two-electron channel in the high pTV region from CMS. In these plots there are excesses in the data over the predicted background yield at the high BDT region. To extract the Higgs signal, global fits are performed on the BDT output distributions of the signal region, and on the other kinematic distributions of the control regions. The extracted observed and expected significance from both experiments, for each leptonic and combined channel, are shown in Table 1. Both experiments observed evidence of Hbb̄ in the VH associated production channel with significance larger than three standard deviations. The measured signal strength values of each lepton channel and the combined signal strength values from both experiments, are shown in Fig. 13. The measurements are for mH=125 GeV (125.09 GeV) for ATLAS (CMS). The measurements are compatible with the SM.



Fig. 12: BDT output distributions from the signal region. (Left) 0-lepton 2-jet from ATLAS. (Middle) 1-lepton 2-jet from ATLAS. (Right) 2-ee high pTV from CMS.

Lepton Channel

ATLAS Obs (Expt)

CMS Obs (Expt)

0
1
2
Combined

0.5 (1.7)
2.3 (1.8)
3.6 (1.9)
3.5 (3.0)

0.0 (1.5)
3.2 (1.5)
3.1 (1.8)
3.3 (2.8)

Table I: The extracted observed and expected significance of the VH signal from both experiments, for each leptonic and combined channel.

Both experiments validate their MVA analysis technique by performing a search for WZlvbb and ZZllbb, vvbb, with one of the Z boson decays into a pair of b-quarks. The reason for choosing these production and decay channels is because they have similar final state signatures as the Higgs signal, and their total production cross section is about an order of magnitude larger than the Higgs signal that the experiments are searching for. From ATLAS, the extracted observed (expected) significance for the combined WZ and ZZ signal is 5.8 (5.3) standard deviations. The combined signal strength is μWZ+ZZATLAS=1.11+0.25-0.22. CMS observed an excess of 5.0 standard deviations with an expected significance of 4.9. The combined signal strength is μWZ+ZZCMS=1.02 ± 0.22.

The WZ and ZZ measurements from ATLAS and CMS agree well with the SM prediction.





Fig. 13: Signal strength parameters of each leptonic and combined channels from MVA analyses of ATLAS and CMS.

To cross check the Hbb̄ search result, a second analysis is performed by ATLAS, which uses the mbb distribution as the main discriminant. Additional selection cuts are applied to optimize the signal sensitivity. The combined mbb distribution, from the three leptonic channels, after subtracting all predicted background contributions except the diboson WZ and ZZ productions, is shown in Fig. 14. The contributions are weighted by their respective values of S/B, where S (B) is the fitted signal (background) yield. The contributions from WH and ZH signals are scaled by the measured combined signal strength value. A clear peak of the diboson can be seen at around mbb~90 GeV, and the shoulder at around mbb~120 GeV indicates the presence of the Higgs signal. For this second analysis, the observed (expected) significance is 3.5 (2.8) standard deviations. The fitted signal strength is μ=1.30+0.28-0.27(stat.)+0.37-0.29(syst.) (at mH=125 GeV).

These Run 2 search results from ATLAS and CMS, based on the MVA technique, are combined with their Run 1 search results. For ATLAS, the combined signal strength is μRun1+Run2ATLAS=0.90+0.28-0.26 (at mH=125 GeV), and the combined observed (expected) signal significance is 3.6 (4.0) standard deviations. For CMS, the combined signal strength is μRun1+Run2CMS=1.06+0.31-0.29 (at mH=125.09 GeV), and the combined observed (expected) signal significance is 3.8 (3.8) standard deviations. This latest search results from ATLAS and CMS show evidence for a SM Higgs boson produced in WH and ZH associated channel, and with the Higgs boson decaying into a pair of b-quarks.





Fig. 14: mbb distribution of the cross check analysis from ATLAS that uses the mbb distribution as the main discriminant.

SUMMARY

Many exciting SM Higgs boson measurements have been performed on the larger Run 2 data samples collected in 2015 and 2016 by the ATLAS and CMS experiments. The experiments improve the precision of measuring the SM Higgs boson properties, and perform more differential measurements of the Higgs boson in the Hγγ and HZZ→4l decay channels. CMS has observed the Hτ+τ- decay within a single experiment, and both ATLAS and CMS have observed evidence of Hbb̄ decay in the VH associated channel. Both experiments have only analyzed the data of the first half of the Run 2 data taking program. We expect more data will be collected during the second half of Run 2 (2017, 2018). Thus, the door is open to explore the possibility of even more interesting discoveries through the LHC experiments.

References

[1] ATLAS Collaboration, JINST 3, S08003 (2008).
[2] CMS Collaboration, JINST 3, S08004 (2008).
[3] L. Evans and P. Bryant (editors), JINST 3, S08001 (2008).
[4] ATLAS Collaboration, Phys. Lett. B 716, 1 (2012).
[5] CMS Collaboration, Phys. Lett B 716, 30 (2012).
[6] ATLAS and CMS Collaboration, Phys. Rev. Lett. 114,191803 (2015).
[7] ATLAS and CMS Collaboration, JHEP 08, 045 (2016).
[8] ATLAS Collaboration, ATLAS-CONF-2017-045, http://cds.cern.ch/record/2273852.
[9] CMS Collaboration, CMS-PAS-HIG-16-040, http://cds.cern.ch/record/2264515.
[10] ATLAS Collaboration, ATLAS-CONF-2017-043, http://cds.cern.ch/record/2273849.
[11] CMS Collaboration, arXiv:1706.09936 (submitted to JHEP).
[12] ATLAS Collaboration, Phys. Rev. D 90, 112015 (2014).
[13] CMS Collaboration, Eur. Phys. J. C. 74, 3076 (2014).
[14] ATLAS Collaboration, Phys. Rev. D 91, 012006 (2015).
[15] CMS Collaboration, Phys. Rev. D 89, 092007 (2014).
[16] ATLAS Collaboration, ATLAS-CONF-2017-047, http://cds.cern.ch/record/2273854.
[17] ATLAS Collaboration, ATLAS-CONF-2017-046, http://cds.cern.ch/record/2273853.
[18] ATLAS Collaboration, Phys. Lett. B 119, 68 (2014).
[19] ATLAS Collaboration, Phys. Rev. Lett. 119, 051802 (2017).
[20] ATLAS Collaboration, JHEP 04, 117 (2015).
[21] CMS Collaboration, JHEP 05, 104 (2014).
[22] CMS Collaboration, arXiv:1708.00373 (submitted to Phys. Lett. B).
[23] L. Bianchini, J. Conway, E. K. Friis, and C. Veelken, J. Phys. Conf. Ser. 513, 022035 (2014).
[24] CMS Collaboration, arXiv:1709.05543 (submitted to PRL).
[25] ATLAS Collaboration, JHEP 11, 112 (2016).
[26] CMS Collaboration, Phys. Rev. D 92, 032008 (2015).
[27] ATLAS Collaboration, ATLAS-CONF-2016-063, http://cds.cern.ch/record/2206201.
[28] ATLAS Collaboration, JHEP 01, 069 (2015).
[29] CMS Collaboration, Phys. Rev. D 89, 012003 (2014).
[30] ATLAS Collaboration, arXiv:1708.03299 (accepted by JHEP).
[31] CMS Collaboration, arXiv:1709.07497 (submitted to Phys. Lett. B).
[32] T. Aaltonen et al., arXiv:1107.3026.

 

Song-Ming Wang received his PhD from the Department of Physics at the University of Iowa, USA. He is now an associate research fellow at the Institute of Physics, Academia Sinica,Taiwan. His current research work is in the ATLAS experiment, at CERN.