1
00:00:00,000 --> 00:00:07,440
Review of Susy searches and Alexis Kalogeropoulos 
will be giving this talk. So I

2
00:00:07,590 --> 00:00:12,750
hand it over to you, Alexis, you have 24
minutes and I will remind you five minutes

3
00:00:12,750 --> 00:00:13,560
before the end.

4
00:00:15,240 --> 00:00:19,740
Okay, so thank you for the introduction.
So this is the experimental SUSY

5
00:00:19,740 --> 00:00:26,220
overview for the Atlas CMS and LHCb
collaborations. So let me start by

6
00:00:26,220 --> 00:00:31,500
reminding you why we need supersymmetry or
SUSY first of all. Well, we need it because

7
00:00:31,530 --> 00:00:36,630
simply put it is one of the most promising
theories we have, beyond the Standard Model

8
00:00:36,780 --> 00:00:41,640
theories, because as we know, Standard Model 
is not the final answer. 

9
00:00:41,640 --> 00:00:49,440
SUSY has many nice features, for example, it
helps solving the hierarchy problem. And

10
00:00:49,470 --> 00:00:54,210
because it allows for a Higgs boson without
fine tuning, now for the experimental

11
00:00:54,210 --> 00:00:59,010
point of view, it is also very appealing
theory because it provides us with a

12
00:00:59,010 --> 00:01:02,910
plethora of possible signatures and
scenarios let alone the excitement if

13
00:01:02,910 --> 00:01:06,630
SUSY were to be found that we would
nearly double the standard model known

14
00:01:06,630 --> 00:01:11,100
particles. Also among other things, SUSY
can provide a massive Dark Matter

15
00:01:11,100 --> 00:01:16,440
candidate, the lightest superpartner or
LSP, if a new symmetry or parity is

16
00:01:16,440 --> 00:01:21,060
invoked, this would basically result in
SUSY particles produced in pairs and also

17
00:01:21,060 --> 00:01:27,930
stable LSP which will escape direct
detector detection. So, briefly SUSY

18
00:01:27,930 --> 00:01:31,590
searches can be categorized in the
following: strong production covers a

19
00:01:31,590 --> 00:01:35,850
rather broad class of inclusive searches
and topologies that benefit from the high

20
00:01:35,850 --> 00:01:39,660
production cross section and as such we
have already set very strong limits on

21
00:01:39,660 --> 00:01:43,410
gluinos and squarks. Then we have
searches for third generation squark pair

22
00:01:43,440 --> 00:01:47,850
production, which are a bit more focused
and most if not all of them have been

23
00:01:47,850 --> 00:01:52,470
motivated from naturalness arguments and
recently we have been able to tackle very

24
00:01:52,470 --> 00:01:56,070
different corners of the phase space
like the top corridor and compressed

25
00:01:56,070 --> 00:02:00,720
regions. Then we have electroweak searches we
mostly talk here about neutralino

26
00:02:00,720 --> 00:02:04,500
chargino pair production. What are the
main challenges is that one deals with

27
00:02:04,500 --> 00:02:10,980
softer kinematics and very often targeting
small mass splittings. In some cases like

28
00:02:10,980 --> 00:02:15,210
searches for direct staus though, we have
just been able to get sensitive to those

29
00:02:15,210 --> 00:02:22,230
models. Finally, we have long-lived particle
searches where we're interested into

30
00:02:22,230 --> 00:02:27,750
displaced jets and leptons with disappearing
tracks and delayed photons. And also other

31
00:02:27,750 --> 00:02:32,520
models, RPV models, which in many cases
overlap with other bsm analysis, but

32
00:02:32,520 --> 00:02:37,710
actually this will be covered on the
plenary on Saturday which I invite you all

33
00:02:37,860 --> 00:02:42,780
of you to attend. Of course I don't have
the time to cover all of this. So today I

34
00:02:42,780 --> 00:02:48,420
will focus mostly on new full run two
results. But there are also four very nice

35
00:02:48,420 --> 00:02:53,790
parallel talks covering in more details
electroweak soft SUSY third generation

36
00:02:53,790 --> 00:02:58,950
and RPV searches. Further
although LHCb to be has no recent SUSY

37
00:02:58,950 --> 00:03:02,940
results, we should not forget some very
nice older ones.

38
00:03:04,680 --> 00:03:10,680
Okay, so as an experimentalist, what we do
here is how to put our hands on SUSY. One

39
00:03:10,680 --> 00:03:15,420
of the biggest problems is that SUSY comes
with relatively small cross sections compared

40
00:03:15,420 --> 00:03:19,470
to the Standard Model processes. But on
the other hand, from a phenomenological

41
00:03:19,470 --> 00:03:24,480
point of view, we actually expect final
states with large missing transverse energy

42
00:03:24,510 --> 00:03:29,430
MET, objects with high transverse
momentum uncertainty, the presence of many

43
00:03:29,430 --> 00:03:35,250
jets and many b quark jets as well. So,
over the last years, a lot of effort have

44
00:03:35,250 --> 00:03:39,060
been put and a lot of kinematic variables
have been proposed, which actually exploit

45
00:03:39,060 --> 00:03:43,890
these features. And a very typical case
which you can see on the top right plot

46
00:03:43,920 --> 00:03:50,190
is the transverse mass. Where due to the
presence of extra MET in RPC scenario signal

47
00:03:50,220 --> 00:03:56,070
in the semi-leptonic decays expected
to populate the tails. And but we expect

48
00:03:56,070 --> 00:04:02,190
the t-tbar actually to have a kinematic
endpoint around the W mass. Further, searches

49
00:04:02,190 --> 00:04:06,660
typically form many signal regions around
the search variables to optimize the

50
00:04:06,660 --> 00:04:13,590
sensitivity and sometimes, in many cases,
a further optimization is achieved by using also

51
00:04:13,590 --> 00:04:18,180
other objects of the event like the number
of leptons and so on. All in all one

52
00:04:18,180 --> 00:04:22,650
can say that the strategy is to surprise
overwhelming background processes as much

53
00:04:22,650 --> 00:04:27,480
as possible and this is done with several
techniques some of more simple some of

54
00:04:27,480 --> 00:04:32,100
these more complex. So, basically we can
categorize background to irreducible

55
00:04:32,100 --> 00:04:36,480
sources and that means, that we can have
similar to the signal experimental

56
00:04:36,480 --> 00:04:42,270
signature like presence of genuine MET,
same number of leptons jets etc. and

57
00:04:42,330 --> 00:04:47,550
simulation is used for the shape and the
normalization is extracted from orthogonal

58
00:04:47,730 --> 00:04:53,580
control regions. Then we have reducible
sources, which are mainly associated to

59
00:04:53,580 --> 00:04:58,560
mis-construction effects like fake
leptons, mis-constructed met and so on

60
00:04:58,560 --> 00:05:03,240
Mismeasured MET, charge flips. And they are
estimated from data because we don't trust

61
00:05:03,240 --> 00:05:08,100
Monte Carlo that much and validated
in validation regions. In the end, the

62
00:05:08,100 --> 00:05:11,940
signal regions which are designed exactly
to be sensitive to the presence of signals

63
00:05:12,510 --> 00:05:17,640
are used to extract the signal. And in the
absence of significant excess, we set

64
00:05:17,640 --> 00:05:22,620
upper limit to the production cross section of
the target models. Now, speaking of

65
00:05:22,620 --> 00:05:27,660
interpretation and models, of course, we
cannot possibly design and execute an

66
00:05:27,660 --> 00:05:32,040
analysis which would have many free
parameters like it happens. We have more

67
00:05:32,040 --> 00:05:37,980
than 100 free parameters, even the
simplest full Susy models. So what we do,

68
00:05:38,070 --> 00:05:43,380
we make some assumptions. And we define
simplifications resulting to a simple set

69
00:05:43,380 --> 00:05:47,610
of simplified models, as we call them
simplified models of supersymmetry, that

70
00:05:47,610 --> 00:05:51,330
basically come the main characteristic is
that we have a very limited number of free

71
00:05:51,330 --> 00:05:56,220
parameters. A typical example is the
production of SUSY particles decaying to lighter

72
00:05:56,220 --> 00:06:00,960
particles, and the only free parameters
are exactly the masses of the mother and

73
00:06:00,960 --> 00:06:05,640
the daughter particles and in many cases,
we also connect those two. That the

74
00:06:05,640 --> 00:06:12,270
daughter particle is half for instance
the the mother particles. We also

75
00:06:12,270 --> 00:06:16,440
regularly assume that the other sparticles
are decoupled, are too heavy and can be

76
00:06:16,440 --> 00:06:22,710
decoupled. Now, one side we know is that
typical we assume the ratios are low along the

77
00:06:22,710 --> 00:06:27,960
decay cascades for these particles. In real
models, this is not usually the case and

78
00:06:27,960 --> 00:06:32,520
the practical impact is that maybe we are
excluding more than we should have been

79
00:06:32,520 --> 00:06:39,780
excluding if we would have been considering real
rather than simplified models. Now, one

80
00:06:39,810 --> 00:06:43,500
very important thing to remember, however,
while doing a real analysis is that we

81
00:06:43,500 --> 00:06:47,970
have to deal with detector imperfections
and ageing effects, which can complicate

82
00:06:47,970 --> 00:06:52,800
our analysis and our life because you have
millions of constructed objects that can

83
00:06:52,800 --> 00:06:57,420
survive the analysis cats. So, for that we
have invested a lot of effort

84
00:06:57,450 --> 00:07:02,670
understanding the detectors and this is
not trivial at all. In CMS, for example,

85
00:07:02,850 --> 00:07:07,020
we compile a list of spurious events which
we can then exclude from the offline

86
00:07:07,020 --> 00:07:12,180
analysis. You can see such a nice example
I think at the bottom plot on the left,

87
00:07:12,210 --> 00:07:18,660
you see how a problematic event in the CMS
ECAL endcap would result in fake

88
00:07:18,660 --> 00:07:24,300
reconstructive MET. And on the right hand
you can see the effect of applying those

89
00:07:24,300 --> 00:07:29,730
noise reduction filters on data. You see
how much the tails are affected on MET and

90
00:07:29,730 --> 00:07:34,320
this is very important because not forget
that missing transparent energy is a prime

91
00:07:34,320 --> 00:07:36,060
object for SUSY sources.

92
00:07:37,410 --> 00:07:42,990
Now, let's start with the analysis review.
So we start with gluino strong production.

93
00:07:43,020 --> 00:07:47,160
Gluino in the first and second
generation squark searches, typically we

94
00:07:47,160 --> 00:07:51,990
expect large met, many jets and many b quark 
jets, whereas typically search

95
00:07:51,990 --> 00:07:56,310
variables try to capture the hadronic
activity and the scale of the event like

96
00:07:56,340 --> 00:08:01,830
HT which is the scalar sum of the jets
of the event or another another variable

97
00:08:01,830 --> 00:08:08,940
that we will see often today, large mass
radius where you see that the mass appears

98
00:08:08,940 --> 00:08:15,870
very nicely in the tails. So, I start
with a very new CMS result this is a brand

99
00:08:15,870 --> 00:08:21,270
new CMS result focusing on gluino decays
to Z bosons and the LSP. Incorporating

100
00:08:21,300 --> 00:08:25,500
also the next to the lightest SUSY particle, 
neutralino two. Neutralino one and 

101
00:08:25,500 --> 00:08:30,540
Neutrino two are mixed states of
neutral Higgs and gauge bosons and also

102
00:08:30,540 --> 00:08:34,440
this analysis specifically considers a
small mass splitting between the gluino

103
00:08:34,440 --> 00:08:39,690
and the neutralino two to be 50
GeV. And also the neutralino one is very

104
00:08:39,750 --> 00:08:45,120
light and is set at one GeV. And that matters
because basically that results into final

105
00:08:45,120 --> 00:08:50,340
states with large MET along with very
energetic Z bosons and the additional

106
00:08:50,340 --> 00:08:55,110
soft quarks coming from the Z decays can
be actually contained in a single

107
00:08:55,110 --> 00:08:59,970
reconstructed jet within a large
radius. Now control regions and search bins

108
00:09:00,000 --> 00:09:05,490
are defined in bins of this MJ variable for
the leading and the trailing jet. And you can

109
00:09:05,490 --> 00:09:08,700
see the definition of the signal regions
and the control regions on the bottom

110
00:09:08,700 --> 00:09:12,960
left. The control regions with a dark blue
are used to estimate the background

111
00:09:12,960 --> 00:09:17,820
normalization. And the light blue is used
to extract the MET say, based on the

112
00:09:17,820 --> 00:09:23,460
assumption that basically MET and MJ
should have minimal correlation. Now,

113
00:09:23,460 --> 00:09:28,770
bottom right, what we can see it's a very
nice sane fit performed on this dark blue

114
00:09:29,070 --> 00:09:35,010
side band regions. When we actually
require the leading jet to lie outside and

115
00:09:35,010 --> 00:09:41,790
the subleading to lie inside the Z mass
window and you can see how nicely the

116
00:09:41,820 --> 00:09:47,910
signal would peak inside the Z mass
window see the signal region. Slide nine

117
00:09:47,940 --> 00:09:52,920
on the left side you see the MET on the
signal region and the upper limits on the

118
00:09:52,920 --> 00:09:57,000
production cross section on the considered
model excluding nicely gluinos

119
00:09:57,000 --> 00:09:59,970
up to 1.5-1.9 TV

120
00:10:01,800 --> 00:10:03,030
Now we're going to

121
00:10:04,800 --> 00:10:09,090
an ATLAS result, an all hadronic with multi
jets and b-tagged jets, hadronic analysis 

122
00:10:09,090 --> 00:10:14,040
from ATLAS that considers also a non-zero
RPV couplings model with a gluino mediated

123
00:10:14,040 --> 00:10:18,660
top squark production. And where
actually baryon number violating

124
00:10:18,660 --> 00:10:22,680
interactions are allowed and stops that
means that can decay to strange and

125
00:10:22,680 --> 00:10:29,010
bottom or to down and bottom quarks. 
MJ again and MET significance are the

126
00:10:29,010 --> 00:10:32,670
main handles to reject background. And
since we are dealing with MET significance

127
00:10:32,670 --> 00:10:37,290
for the first time in this talk, so the MET
significance in particular, it's a variable

128
00:10:37,320 --> 00:10:43,680
that tests in particular the hypothesis,
how compatible the total transverse momentum

129
00:10:43,680 --> 00:10:49,260
is with non interacting particles, as
basically it quantifies by how much the

130
00:10:49,260 --> 00:10:54,900
measurement can be associated to possible
mis-measurements. So, large jets

131
00:10:54,930 --> 00:11:00,840
large values of this MET significance is
linked to non interacting particles. And in

132
00:11:00,840 --> 00:11:04,710
this analysis actually what is also very
nice is that MET significance has been

133
00:11:04,710 --> 00:11:10,200
optimized to capture the response of the
ATLAS detector. The main backgrounds are

134
00:11:10,230 --> 00:11:14,580
multijet through QCD processes for
instance, when we have semileptonic B and C

135
00:11:14,580 --> 00:11:19,560
hadron decays and it is estimated from a
lower jet multiplicity control region by

136
00:11:19,560 --> 00:11:24,060
building MET significance templates,
because we assume that this should be

137
00:11:24,060 --> 00:11:30,420
mostly uncorrelated with MJ. And on
the right you can see a very nice closure

138
00:11:30,420 --> 00:11:38,940
test on the MET significance with the different
processes populate the plot. Okay, so that

139
00:11:38,940 --> 00:11:42,750
brings me to the results were on top left
you can see the yields in one of the signal

140
00:11:42,750 --> 00:11:47,370
regions. And on the bottom you see the
results on this note one of the nonzero

141
00:11:47,370 --> 00:11:53,970
RPV coupling model where the reach is just
above 1.5 GeV for stop masses of 400 GeV

142
00:11:54,330 --> 00:11:59,580
and it extends considerably the
previous result. On the right the

143
00:11:59,580 --> 00:12:03,270
potential chargino in a mediated
simplified model compared on the bottom

144
00:12:03,270 --> 00:12:08,220
with some recent CMS result and as you can
see both of them have about the same

145
00:12:08,220 --> 00:12:12,960
sensitivity excluding about two TeV for
gluino masses for for light

146
00:12:12,990 --> 00:12:13,830
neutralinos.

147
00:12:15,210 --> 00:12:15,780
Okay,

148
00:12:17,190 --> 00:12:22,680
now another CMS result in one lepton
final state. The experimental signature

149
00:12:22,680 --> 00:12:28,050
here is one lepton, multijets, B tag and
large MET again, but also this

150
00:12:28,050 --> 00:12:32,550
analysis exploits the presence of initial
state radiation jets and the main source

151
00:12:32,550 --> 00:12:39,120
variables are the transverse mass, and the MJ.
Now, a property of MJ that is used in this

152
00:12:39,120 --> 00:12:43,020
analysis is that for t-tbar events
typically, which can have a large jet

153
00:12:43,020 --> 00:12:47,640
multiplicities and MJ is mostly
uncorrelated with the transverse mass

154
00:12:47,670 --> 00:12:54,720
meaning that the t-tbar at the high MT can
be estimated from a MJ distribution

155
00:12:55,110 --> 00:13:00,330
from a low MT control region. There are
several SRs and control regions being

156
00:13:00,360 --> 00:13:07,080
in MJ, MT, MET, multiplicity of jets and
B tagged jets and the backgrounds is estimated

157
00:13:07,080 --> 00:13:12,360
from control regions again under the
assumption that MJ and MT are correlated. In

158
00:13:12,360 --> 00:13:16,920
any case potential correlation is taken
into account by a correlation factor K,

159
00:13:16,950 --> 00:13:24,810
which as you can see on the bottom right
Plot is very close to unity. Now, on top

160
00:13:24,810 --> 00:13:29,070
half of the page, you can see the MJ
distribution in one of the signal regions

161
00:13:29,070 --> 00:13:33,240
because there are more than one signal
regions and at the bottom left hand the

162
00:13:33,240 --> 00:13:40,110
yields for it. The results are independent
with the use of simplified models and

163
00:13:40,110 --> 00:13:46,230
masses up to 2.2 TV for gluinos are
excluded for this T1tttt model, which is

164
00:13:46,260 --> 00:13:52,200
this limit is among the strongest limits
we have for gluinos. Okay, so

165
00:13:52,230 --> 00:13:57,150
overviewing the gluino searches one can
say that the Run two has put very stringent

166
00:13:57,150 --> 00:14:02,250
limits on gluino masses, even more than
two TeV gluinos seem to be well

167
00:14:02,250 --> 00:14:06,780
excluded and typically we have weaker
limits for chargino and boson mediated

168
00:14:06,780 --> 00:14:11,490
decays because we have softer decay
products, but still these limits sit well

169
00:14:11,490 --> 00:14:16,380
above one TeV. On the other hand, we expect
to reach even above three TeV at the high

170
00:14:16,380 --> 00:14:24,090
luminosity LHC. Switching gears now, we go
to stop searches where basically when can

171
00:14:24,090 --> 00:14:27,810
divide the phase space as a function of the
mass difference between the stop and the

172
00:14:27,810 --> 00:14:32,220
neutralino, and that basically defines
what kind of decays are kinematically

173
00:14:32,220 --> 00:14:38,250
allowed and as we can have two three or
even four body decays. Of very, of

174
00:14:38,250 --> 00:14:42,000
particular interest is the top corridor
where basically the mass difference

175
00:14:42,000 --> 00:14:46,500
between the stop and the neutralino is
about the top mass meaning that stops are

176
00:14:46,500 --> 00:14:52,980
produced at rest. And this is very, very
difficult to do analysis there. And the 

177
00:14:52,980 --> 00:14:56,760
t-tbar is the dominant background of course,
now once again we have variables which try

178
00:14:56,760 --> 00:15:02,430
to distinguish potential signal against
the background. For example top-ness which

179
00:15:02,460 --> 00:15:08,430
differentiates through two-lepton dileptonic
events. Of course we have many more

180
00:15:08,550 --> 00:15:14,190
In use transverse mass and MT 2 and so
on. And with all of them try to exploit

181
00:15:14,220 --> 00:15:18,720
the presence of extra MET in signals
basically the characteristics peculiar

182
00:15:18,750 --> 00:15:24,120
the particular characteristics that signal
have. Okay, so first of all, I'll be

183
00:15:24,120 --> 00:15:29,550
showing hadronic stop analysis from Atlas
that covers all of these two three and

184
00:15:29,550 --> 00:15:35,130
four body the case as we can see on the
top left plot, what is very nice in this

185
00:15:35,130 --> 00:15:38,940
analysis is that different regions were
optimised with different variables like

186
00:15:38,940 --> 00:15:45,900
the MET significance, MT, but also they made
use of ISR sensitive variables, which

187
00:15:46,140 --> 00:15:51,660
would help you significantly on the top
corridor and basically this is formed as

188
00:15:51,660 --> 00:15:57,360
the ratio, just for reference, of MET divided by
the PT of the ISR. While they also use

189
00:15:57,360 --> 00:16:02,100
the soft PT in the soft PT region they
use also other angular variables, in order to reject

190
00:16:02,130 --> 00:16:07,650
mis-measurement and B soft tagging techniques
as well. I think this is just a very good

191
00:16:07,650 --> 00:16:12,720
example where actually you need it shows
me that we need dedicated optimization if

192
00:16:12,720 --> 00:16:16,650
we want to reach this difficult
corners of the phase space and

193
00:16:16,710 --> 00:16:20,340
different standard model processes
dominate on different signal regions and

194
00:16:20,340 --> 00:16:25,260
they are estimated from data in orthogonal
control regions. Now, coming to the

195
00:16:25,260 --> 00:16:29,730
results on the bottom half, you can see
the protections as a function of stop

196
00:16:29,730 --> 00:16:35,880
versus neutralino and we exclude stop
masses for about 1.2 TeV for massless LSP

197
00:16:36,390 --> 00:16:42,810
extending by about 200 GeV previous
results. CMS has also the same sensitivity

198
00:16:43,140 --> 00:16:47,250
from another analysis although this
particular CMS result analysis was not

199
00:16:47,250 --> 00:16:49,590
optimized for three- and four-body decays.

200
00:16:51,420 --> 00:16:59,130
Okay, now coming to yet another one lepton 
stop analysis results from ATLAS. This

201
00:16:59,130 --> 00:17:05,220
analysis employs both cut-and-count and
shape analysis to enhance sensitivity. And

202
00:17:05,220 --> 00:17:09,270
again typically we have many signal
regions optimised after applying several

203
00:17:09,270 --> 00:17:13,830
cuts on sensitive variables, dealing
now with backgrounds among the most

204
00:17:13,830 --> 00:17:17,100
important sources are mis-measurements
and sources that contribute to lost

205
00:17:17,100 --> 00:17:22,680
lepton, meaning lost lepton is when you
have a one of the leptons misidentified or

206
00:17:22,830 --> 00:17:28,830
mis-reconstructed and for that, the top-ness
is used which basically quantifies how

207
00:17:28,830 --> 00:17:34,440
well an event can be reconstructed under
the hadronic top hypothesis. And what

208
00:17:34,440 --> 00:17:39,780
is also very nice is that there is also a
similar CMS one lepton analysis, which

209
00:17:39,780 --> 00:17:43,650
also use top-ness as discriminating
variable for this kind of background.

210
00:17:44,550 --> 00:17:48,750
Upper limits are presented on the bottom
half of the page and once more we say that

211
00:17:48,750 --> 00:17:57,150
stops above one TeV are well excluded.
Coming now to a very to a CMS result which

212
00:17:57,150 --> 00:18:01,170
is also very new. It was released this
week targeting two lepton final

213
00:18:01,170 --> 00:18:06,870
states. And among other models it
considers chargino mediated decays. The main

214
00:18:06,870 --> 00:18:12,090
variables are MT two and a variant of it
which also incorporates the multiplicity

215
00:18:12,090 --> 00:18:16,800
the presence of b tagged jets of the event. Now,
the key idea behind the MT two is that

216
00:18:16,800 --> 00:18:21,990
actually, you expect them to have a
kinematic endpoint for background events

217
00:18:22,020 --> 00:18:27,720
even if you have two neutrinos in the event.
The MET significance again is used here to

218
00:18:27,720 --> 00:18:31,770
suppress mis-reconstructed particles,
which can be associated for instance, with

219
00:18:31,770 --> 00:18:37,620
pileup interactions, which is the main
source of mis-reconstructed MET. And

220
00:18:37,710 --> 00:18:43,470
just to give you some more information
that this follows a chi square

221
00:18:43,470 --> 00:18:48,450
distribution with two degrees of freedom
for events with no genuine MET. Now,

222
00:18:48,450 --> 00:18:53,160
right hand shows the distribution of the
MET significance in a Drell-Yan 

223
00:18:53,160 --> 00:18:56,850
region, and I think that nicely
demonstrates how different backgrounds

224
00:18:56,850 --> 00:19:02,310
contribute while worth mentioning that The
main background source comes from mismeasured

225
00:19:02,310 --> 00:19:09,300
jets that could enter your MT2 tail
distribution in the signal region. Okay, slide

226
00:19:09,300 --> 00:19:15,030
19. That brings me to the result of this
analysis. On the left you see the yields

227
00:19:15,060 --> 00:19:19,110
in the control and the signal regions,
which are fit simultaneously to extract

228
00:19:19,110 --> 00:19:22,950
the signal. And the absence of any
significant excessive results are

229
00:19:22,950 --> 00:19:28,830
interpreted in simplified models, one of
which is shown on the right. And this

230
00:19:28,830 --> 00:19:33,690
extends we have stop production to top
and a neutralino decays and this

231
00:19:33,690 --> 00:19:41,730
extends previous CMS result by about 125 GeV. 
Coming back to Atlas we have also a

232
00:19:41,730 --> 00:19:47,430
new result was released a couple of days
ago, which was present.

233
00:19:48,810 --> 00:19:52,230
Yes, apologize for interrupting you. You
have four minutes left.

234
00:19:52,830 --> 00:19:56,790
Okay, thank you. So this analysis
considers the presence of on-shell

235
00:19:56,790 --> 00:20:00,600
Higgs and Z bosons and once again, we
found the transverse mass, MET

236
00:20:00,600 --> 00:20:05,880
significance and a multiplicity of jets and 
b-tagged jets. The analysis considers also

237
00:20:05,880 --> 00:20:09,720
three and one lepton final states, each
one of these final states with different

238
00:20:09,720 --> 00:20:14,820
backgrounds, you can see a very nice
closure test on the top of the bottom left

239
00:20:14,820 --> 00:20:21,240
on the PT Z-enriched region and the
interpretation or one of the considered

240
00:20:21,240 --> 00:20:25,680
models is also shown. And this actually
demonstrates a very nice improvement

241
00:20:25,680 --> 00:20:29,490
compared to the previous results
practically extending results between

242
00:20:29,520 --> 00:20:37,560
hundred and 300 GeV depending on ??? modes
Now another new ATLAS result which was

243
00:20:37,560 --> 00:20:42,510
released recently considers multi b=jets
in RPV scenarios, given that it

244
00:20:42,510 --> 00:20:46,620
considers high number of jets and beta
jets. The main purpose of this study is

245
00:20:46,620 --> 00:20:53,640
estimated from data. And what is very nice
with this analysis it is actually that's

246
00:20:53,640 --> 00:20:58,920
the first analysis that sets limits are to
stop decaying exclusively to chargino and a

247
00:20:58,920 --> 00:21:04,020
B-quark. So that brings me to the overview
slide for stop searches where as you can

248
00:21:04,020 --> 00:21:11,490
see from the summary plots from CMS top quarks
and Atlas, we have excluded

249
00:21:11,490 --> 00:21:16,050
stops lighter than 1.2 TeV for low
neutralino masses, the limits are weaker

250
00:21:16,260 --> 00:21:20,670
along the diagonal because we have softer
objects. One thing to mention is that both

251
00:21:20,670 --> 00:21:24,390
collaborations incorporate more and more
elaborate techniques and variables in

252
00:21:24,390 --> 00:21:30,210
order to increase the rates. Further
bottom right, you can see that the high

253
00:21:30,210 --> 00:21:36,180
luminosity LHC we can expect to
maybe reach up to two TeV stops. Now going

254
00:21:36,180 --> 00:21:40,950
to electroweakino searches now,
they pose a very specific challenge because

255
00:21:40,950 --> 00:21:45,270
of this small production cross section and the
soft kinematics that we have. But on the

256
00:21:45,270 --> 00:21:49,290
other hand, we have very clear signatures
like MET, multi leptons and other handles.

257
00:21:50,250 --> 00:21:53,460
Of course, we can talk about many
different scenarios on searches here like

258
00:21:53,460 --> 00:21:57,330
the gaugino neutralino, gauge mediated and
so on and so forth. But what I will

259
00:21:57,360 --> 00:22:01,950
actually do to focus on this talk is about
the compressed regions and we have seen

260
00:22:01,950 --> 00:22:06,090
recently that people are trying to use
handles like vector boson fusion

261
00:22:06,090 --> 00:22:11,430
topologies and exploit the ISR, the
presence of ISR, in the models to increase

262
00:22:11,460 --> 00:22:18,120
sensitivity. So a very nice example of
that both is this CMS analysis for the

263
00:22:18,120 --> 00:22:21,810
gaugino neutralino mediated stau
production looking for very soft taus

264
00:22:22,350 --> 00:22:26,940
optimizes around the presence of ISR.
Without going to too much into details,

265
00:22:26,970 --> 00:22:31,320
the point here is that the star that you
see the point, which is very excluded in

266
00:22:31,320 --> 00:22:37,770
this analysis, actually sits in a region,
which is way before beyond any any

267
00:22:37,800 --> 00:22:42,750
sensitivity than any other chargino neutralino
analyses.  And I think the the

268
00:22:43,080 --> 00:22:47,640
take home message here is that we should
be smart and start using maybe more VBF

269
00:22:47,640 --> 00:22:53,970
and ISR techniques in order to access these
compressed and very difficult regions. So

270
00:22:54,000 --> 00:22:58,260
if one wants to review electroweakino searches,
again, we have very strong limits the

271
00:22:58,260 --> 00:23:02,610
strongest are for slepton mediated decays
Followed by boson mediated decays. On

272
00:23:02,610 --> 00:23:06,600
the right hand, you can see the summary
plots from Atlas and CMS. And I think it's

273
00:23:06,600 --> 00:23:11,850
clear that reaching different corners of
the phase space requires well established

274
00:23:11,850 --> 00:23:18,450
methods done more than just going with a good for
everything model or strategy. So what we

275
00:23:18,450 --> 00:23:22,650
expect in the future, well, given a high
Lumi gives access to models that were

276
00:23:22,650 --> 00:23:26,880
suppressed up to now we can expect that
the high luminosity will LHC will be

277
00:23:26,880 --> 00:23:30,750
very interesting because it will allow us
to have a lot of the phase space, as you

278
00:23:30,750 --> 00:23:34,170
can see on the projections. Of course, the
problem is that these projections have

279
00:23:34,170 --> 00:23:38,490
been made with certain assumptions. But I
think the point here is that it will be

280
00:23:38,730 --> 00:23:43,710
the electroweakino searches will get more
and more attention in the future. And that

281
00:23:44,760 --> 00:23:54,570
the so called CMS, well, it's it's certain
that both ATLAS and CMS have a very rich SUSY

282
00:23:54,570 --> 00:23:59,250
physics program producing very nice
results. Of course, we expect more full

283
00:23:59,250 --> 00:24:04,050
run 2 results. On the making, among other
things, for example, the full likelihoods

284
00:24:04,050 --> 00:24:08,310
being released by Atlas. And don't forget,
as we ramp up and we prepare for Run three,

285
00:24:08,310 --> 00:24:12,540
and looking forward to the high
luminosity LHC, we will benefit from

286
00:24:12,540 --> 00:24:16,650
not only from additional Lumi to tackle
with very difficult corners and models,

287
00:24:17,130 --> 00:24:21,930
while also using more targeted triggers,
sophisticated tools, upgraded and improved

288
00:24:21,930 --> 00:24:26,400
detectors allow us to believe I think that
it will be very exciting in this period,

289
00:24:26,760 --> 00:24:32,160
propelling us to uncharted waters. So
stick around more will come. Thank you for

290
00:24:32,160 --> 00:24:32,640
your time.

291
00:24:35,010 --> 00:24:40,890
Okay, thank you, Alexis for the very nice
overview of Susy searches. So I will now

292
00:24:40,890 --> 00:24:46,620
open the floor for questions. So you can
raise your hand, there's a feature at the

293
00:24:46,620 --> 00:24:52,440
bottom of your screen. If you cannot talk.
You can also type your question in the

294
00:24:52,440 --> 00:25:01,050
chat and we can read that question. So I
don't see any right now. Maybe I will ask

295
00:25:01,140 --> 00:25:08,340
one because I was trying to figure out on
slide 17, and maybe 22. But if you go to

296
00:25:08,340 --> 00:25:18,450
slide 17, there's on the Atlas exclusions,
you see a significant dip. Whereas on the

297
00:25:18,450 --> 00:25:23,100
CMS side, I don't know if we're comparing
apples to apples here, but on the CMS

298
00:25:23,100 --> 00:25:28,170
side, you appear to have searches that
excluded that that dip.

299
00:25:29,790 --> 00:25:31,410
Am I missing something or

300
00:25:31,980 --> 00:25:38,070
object? I think it has to do exactly what
specific optimizations and maybe more

301
00:25:38,070 --> 00:25:43,020
signal regions defined with b-tags and
so on. So I don't think it's something

302
00:25:43,080 --> 00:25:47,940
that it has anything to do with Atlas or
CMS detectors that say other than the

303
00:25:47,940 --> 00:25:50,910
strategy that people follow in different
analysis.

304
00:25:53,040 --> 00:25:59,550
Okay, thank you. I saw there was a raised
hand but then it it went away. So I guess

305
00:25:59,550 --> 00:26:01,290
you You gave him the answer.

306
00:26:04,200 --> 00:26:05,220
I don't think

307
00:26:05,220 --> 00:26:08,160
there was one that is appeared again.
Yeah.

308
00:26:13,110 --> 00:26:16,020
Okay Federico, if you want to comment.

309
00:26:17,430 --> 00:26:20,010
I can you hear me? Yes, go ahead.

310
00:26:20,340 --> 00:26:25,260
Yeah, I think the answer was was okay. I
believe that the main difference in this

311
00:26:25,260 --> 00:26:30,060
region is the trigger strategy that has
been used in the two searches. For

312
00:26:30,060 --> 00:26:35,040
example, in ATLAS, we rely the least
at least for this result on the MET

313
00:26:35,100 --> 00:26:40,260
trigger. So we lose a little bit of
acceptance in this region where the MET

314
00:26:40,260 --> 00:26:46,260
spectrum is not voir dire, while CMS
recovers sensitivity thanks to the single

315
00:26:46,260 --> 00:26:47,310
lepton triggers.

316
00:26:49,800 --> 00:26:50,880
Okay, thank you

317
00:26:55,980 --> 00:26:57,750
don't see other questions.

318
00:26:59,190 --> 00:27:07,260
If we still have Some time I can ask out of
curiousity speaking exactly of

319
00:27:07,260 --> 00:27:11,940
triggers and speaking of the future you showed 
high lumi but in view of the run

320
00:27:12,000 --> 00:27:19,710
three given that maybe we cannot really
just go forward in the just in the mass

321
00:27:19,710 --> 00:27:26,100
reach about focus on this difficult
regions are there may be new strategies

322
00:27:26,100 --> 00:27:32,700
that are being thought maybe for triggers
or specific things that have couldn't

323
00:27:32,700 --> 00:27:35,850
been done now but maybe change of priority.

324
00:27:37,980 --> 00:27:38,760
Say for example,

325
00:27:40,200 --> 00:27:45,960
I could say for instance that one of the
certain challenges and we have to deal

326
00:27:45,960 --> 00:27:51,330
with that is that with higher pileup, you will
have a problem with the MET due to

327
00:27:51,330 --> 00:27:57,600
fakes. Okay, so the triggers there are
pileup dependent, and this is something

328
00:27:57,600 --> 00:28:02,010
that we definitely need if we actually
would need to squeeze out more on the

329
00:28:02,010 --> 00:28:09,510
compressed region. So I think there is
already there already efforts exactly

330
00:28:09,510 --> 00:28:15,690
optimizing the MET triggers, but also
soft lepton triggers because typically the

331
00:28:15,690 --> 00:28:21,600
leptons, the single lepton triggers they
have thresholds above 25 or even 30 GeV.

332
00:28:22,140 --> 00:28:27,150
We have strategy we have triggers, dileptonic
 triggers, which can go lower,

333
00:28:27,390 --> 00:28:30,750
but sometimes this is not enough. You need
to go even even lower. Transcription: F. Blekman