Searches for electroweak production of supersymmetric gauginos and sleptons with the ATLAS detector

Results from the searches for electroweak production of gauginos or sleptons decaying into leptonic final states performed using 20.3 fb −1 of proton-proton collision data at √S = 8 TeV recorded with the ATLAS experiment at the Large Hadron Collider are presented. No significant excesses are observed with respect to the prediction from Standard Model processes. Limits are set on a wide range of SUSY models.


Introduction
Since the discovery of Higgs boson at LHC, the most essential part of standard model(SM) was confirmed.However, we still have some remaining issues after this: hierachy problem, dark matter, origin of gravity and no gauge unification at higher scale.Supersymmetry(SUSY) has become a very appealing extension of SM to answer these questions.Currently, we have set limits of most SUSY models probe masses up to 900GeV(squarks) and 1.4T eV(gluinos).
The cross-sections of different SUSY procedure are shown on Fig. 1.The EWK sector provides a promising approach for new physics among all SUSY searches.The hadronic activity is quite low in a certain EWK process despite the low production cross-section.For most of the EWK models, we have 1-4 leptons in the final state, as well as large missing transverse energy(Etmiss) and 0-2 jets(or b-jets).

Search for direct stau production in events with at least two hadronic taus and missing transverse momentum using multivariate analysis technique
This part is an update of JHEP 1410(2014)96.

direct stau production
The direct stau production(see Fig. 2) have the final states: two opposite-sign hadronically decaying taus, no jet, and large E T miss .Limited sensitivity was observed using cut-and-count analysis due to the low cross-section of direct stau process.Multivariate analysis(MVA) is used to improve this result.We defined one boosted decision tree(BDT) signal region(SR) for this intension, as shown on Fig. 3.

background
The main background of this study includes: -W+jets(1real + 1fake lepton).This background is estimated by a normalization to data in dedicated WCR.
-multi-jet.We apply an "ABCD" estimation to this background source, using tau-ID and m T2 as the two uncorrelated variables, shown in Fig. 4.
-Other sub-dominant background.This is estimated by Monte Carlo(MC) simulation.

results
The results of this direct stau update was shown in Fig. 5. Significant improvement is observed in the low neutralino mass region using MVA compared to cut-and-count analysis.

background
The SR of this analysis includes the following main background: -"prompt" leptons(diboson, H), estimated from MC simulation directly.
The disctributions of E T miss and Mjj are shown on Fig. 7.

results
The fitting results of of this topic was shown on Fig. 8.The best observed upper limit is found for χ 0 1 mass of 120GeV and m( χ ± 1 )-m( χ 0 1 )=25GeV.Slightly stronger sensitivity is also observed for higher χ ± 1 masses.

Search for supersymmetry in compressed scenarios
The compressed scenarios are some SUSY scenarios with small mass difference between sparticles and neutralinos, leading to low-momentum leptons in final states.Generally, low mass-splitting, high Figure 8. 95% CL exclusion limits on the cross-section for VBF production of χ ± 1 χ ± 1 .The limits have been set with respect to the mass difference between the χ ± 1 and χ 0 1 .
soft decay products and SM-like behaviour leads to many difficulties for this analysis.The results of re-optimization of three sub-topics are shown: -two opposite-sign(OS) light leptons.

Two OS light leptons: Event selection and background estimation
The experimental signature of this study is: two OS light leptons and E T miss shown in Fig. 9. Two SRs are defined for this analysis, requiring a high pT initial-state-radiation(ISR) jets, which are sensitive to small and moderate mass splittings.Discriminate variabls "Super razor" are also used in the SR definition, shown on Fig. 10.The main background in the SRs are: -Irreducibel background(WW, top, ZV), estimated by normalized MC simulation in dedicated CRs.
-Reducibel background(all fake sources) -Matrix Method adopted for this background source.Two SRs are defined with low pT leptons to target mass splittings of 4-15 and 15-25GeV, while another two SRs which request a jet with pT>50GeV to target ISR events for both splitting regions.
The MET distribution in the SRs are shown on Fig. 14, with the following SM background: -irreducible(prompt) leptons, including diboson and triboson, estimated from MC simulation.
-reducible(fake) leptons, including Vjets, WW, top and tt, used Matrix Method for the estimation.

results
The fitting results of all the three scenarios are shown on Fig. 15 and Fig. 16.For the OS2l analysis, reoptimized analysis nicely complements the already published one in the region of low mass splittings close to the diagonal.For the SS2l/3l analysis, re-optimized analysis nicely complements the already published one in the region of low mass splittings close to the diagonal.The combination of the new analyses give an improved sensitivity to compressed scenarios up to χ ± 1 masses of 250 GeV.

Figure 1 .
Figure 1.SUSY cross-sections for different production against SUSY particle mass.

Figure 2 .
Figure 2. Feynman diagram of the pair production of charged staus decaying into 2 taus in the final state.

Figure 3 .
Figure 3. Distribution of BDT response and the SR definition.

Figure 4 .
Figure 4. Introduction of "ABCD" estimation.Two uncorrelated variables are used on each axis, defining 3 CRs with enough statistics and 1 SR.The extrapolation from the CRs provides a good estimation to the event yields in SR.

Figure 6 .
Figure 6.Feynman diagram for VBF production of C1C1 with two SS leptons in the final states.

Figure 7 .
Figure 7. Distribution of the E T miss and di-jet invariant mass in SR.

Figure 11 .
Figure 11.Feynman diagram for compressed scenario with two SS leptons in the final states.

Figure 12 .
Figure 12.Distributions of Δφ and m lep1 T in SRs.

Figure 13 .Figure 14 .
Figure 13.Feynman diagram for compressed scenario with 3 leptons in the final states.

Figure 15 .Figure 16 .
Figure15.95% CL exclusion limits on the cross-section for compressed scenario with 2 OS leptons in the final states.

Figure 17 .
Figure17.95% CL exclusion limits on the cross-section for compressed scenario with 2 SS leptons or 3 leptons in the final states.