Higgs decays to ZJPsi, JPsiJPsi and YpsilonYpsilon (FTR-21-009)

CADI: https://cms.cern.ch/iCMS/analysisadmin/cadilines?line=FTR-21-009&tp=an&id=2531&ancode=FTR-21-009

Color Code

Comments from Sezen

L36: in and BSM —> in SM and BSM? add missing SM

Sec 4: Please prepare a table summarizing the selection.

done

Please prepare a dedicated section describing the extrapolation to HL-LHC, in particular how the samples were extrapolated, how xsects are treated, that the fit was used for extrapolating data, but (to my understanding) a new fit was done for HL-LHC, etc.

Clarify early on how the electron channels will be indirectly treated. This only becomes clear in the results section.

After the introduction, add a section describing the strategy, including the addition of the electron channels in Z J/psi.

Figs 2 and 3: “x� missing from the signal legend lines.

Add 'x' to BF value in plot.

In my opinion, it might be nicer to have Figs 2 and 3 in the results section, after everything is described.

done

L227: You can reduce lumi uncertainty to 1%, which is the HL-LHC value.

done

Are there no systematics effecting the background shapes?

The background parameters are floating.

A plot summarizing the Run2, 3000fb-1 and 4000fb-1 upper limits together with systematics would be very good.

Since the limits with respect to the SM in the two channels are quite different, a graphical representation will be difficult. We include a new result table.

For the version after preapproval, let's also change the PAS title to reflect that the analysis is done for HL-LHC. Modifying the first part of the abstract would also be helpful.

done - instead of at CMS changed to .. at HL-LHC

Comments from Sascha

My major comment is that currently the PAS looks like copy of your Run 2 PASes, and it is not clear what was done to get this projection for the HL-LHC. I suggest you to shortened wherever possible the Run 2 and general descriptions and focus more on what you did to get the final results for the HL-LHC

1. Add to the title " at the HL-LHC"

done

abstract last line: reach with CMS detector at the HL-LHC

done

(I dont think you use explicitely the upgraded CMS detctor in this projection, do you?)

We use the new acceptance in tracking, and muon and electron identification

l.25 Figure 1 ...

done

I think that the introduction, even it is nice, is very long. You have referenced 2 PASes that you use from Run 2 and I think you can just refer to them. The most important information that is difficult to separate from this text is that you are talking about 2 decay examples that have very low SM Br , and at Run 2 you can only reach a certain CL and now you try the HL-LHC since this is certainly a place where you can gain a lot. I would suggest to shortend the introduction, refer to HIG PASes and clean up the message.

We shorten the introduction. We would though like to make clear to the reader the connection between the channels in the description in this note, with no dependence on the previous notes.

Section 2 is a little bit confusing. Do you use anything from Phase 2 upgraded detector in this analysis? The description you have is about Phase 1 we need to change it to short description of Phase 2 where you can say at the end that you assume that the major efficiency will stay as they are in Run 2 for this projection. Some short version of this section you can take from FTR-18-019 where we need to add recent TDR references.

We shorten the description. We describe the relevant sub-detectors, most importantly the new acceptance which we include in the projection calculation.

l.136 I dont think you use 200 PU, this is Run 2 sample I think, one needs to say here what you do about 13 to 14 GeV, what is the strategy for projection etc

We include a strategy section. Most importantly, we find no dependence on the PU in previous analyses.

section 4 is once more, more like Phase 1 description that you can just reference.

Updated the section. Include a summary table.

section 6 we need to include YR uncertainties

BR uncertainties are included (from PDG) - systematics vii

ARC Comments from Michele

abstract:

L2: "to search...Yukawa" -> to probe Yukawa

L3: please remove "and add loop diagrams". it is not clear and not fit for the abstract.

L5: "recently,". What do you mean?

L6: "four-lepton": does this include electrons and muons?

L9: "to benchmark" is jargon. Please rephrase.

main document:

L12: "decays ... to mesons provide...": how? is it through loop diagrams? please clarify.

L13: "to study ... BSM physics" is weird. Perhaps "and search for hints of new physics.

L19+1: -> The left diagram

L23: -> two diagrams to the right...

L32: "larger then the expected SM...". It sounds like there is an excess, although the sensitivity of the measurement is far from SM expectations. Please rephrase.

L48: this note -> this document

done

L122: "four muon final states ... is about 13%". Is this value an average of the two separate final states? what are the values for the two final states separately? Also, it should be defined earlier (in the introduction) whether the analysis addresses also electrons or only muons.

13% for all muon final states. For Z->ee (ee or mumu) it is 18%, which is given as additional information. We will clarify this.

L123: "...about 13%". What is the reco/misreco efficiency for the central and for the new added detectors for HL-LHC. It would be good to clarify the assumptions made on the detector efficiencies for barrel/endcaps, also in light of the larger expected PU, and how they are estimated.

We only take into consideration the increased acceptance in eta, and assume that the performance of the detectors is comparable to Run-2 (greater 95% (reco,id) and systematic uncertainties on the efficiencies smaller than 3%).

L123: decaying into e+e- pairs ... is 18%: does it mean the value is the same for 2mu2e or 4e? for both final states? or is this an average? what are the separate values?

The acceptance increases most for the electrons.

L133: here and elsewhere: this note -> this document

done

Table 1: - caption: muons are mentioned. What about electrons?

Electron channels are not explicitly reconstructed here.

- the final state is referred to as "YY" in the text. Please use the same. Also, "n" and "m" are introduced without being defined.

- "Y(mu+mu-)+mu" is not a trigger. Is this a 3 muon trigger? please clarify. It is not clear what is in the rows and columns

This is a 3 muon trigger with the additional requirement that one neutral muon pair falls within a dilepton invariant mass interval 8 - 12GeV. Add text. . - "minimum pT": 1GeV and 5GeV are extremely low. If this is realistic, please explain and add a Ref. Also, what are the eta cuts?

This is the 4-muon pT. In H-Z J/psi this cut is superficial as momenta are higher. We remove it.

L158: "at least four muons": only muons?

At least one candidate. After all selection criteria are applied, only single candidate events survive.

L161: "51 ... events". Is this for Run2 only? please clarify. Also, it is weird that data events are looked at first.

Yes; we describe the strategy first in section 2.

L166: "A Y(nS)...". It would be better to introduce/discuss the two searches clearly, ie an individual paragraph for each.

L161ff - we indeed miss the distinction that here we describe Z J/psi specifics.

L172: Please remove "(left) and 3 (right)"

done

L173: -> for representative simulated samples.

this is: for one representative sample

L174: differences in efficiencies: please clarify the assumptions made on the reco/misreco efficiencies in the text. This is especially needed for the new HL-LHC detectors and conditions.

Now removed this paragraph.

L179: around 20%: how does this correlate to the 13% and 18% that was mentioned earlier? Do we have to assume that the Run2 efficiency was 13%/18% lower than 20%. What was the polarization assumption earlier? Please clarify/rephrase the text.

20% is acceptance times efficiency. It includes the HL-LHC detector acceptance. We removed this paragraph.

L185: "obtained from data" -> "extrapolated from Run2 data" perhaps?

done - obtained from Run-2 data.

L185: possible bias: please be specific what possible bias (due to what) and which alternative functional forms were used.

done

L189: these were referred to as "final states": channel-> final state

Here and elsewhere. It would be clearer to use parentheses: (Z) (J/psi).

L192: in the previous -> in the Run2

done

L195: here and elsewhere. Y pair: It was referred to earlier as "YY". Please change.

done

L204: 1%: add a Ref to justify this assumption or some logic that clarifies why this number was chosen.

added reference [57]

L214: lepton: electrons and muons?

Both, but here specifically muons - changed

L214: -> energy resolution ?

mass resolution

L224: change in detector conditions: please clarify what that means.

Removed.

L228: lepton pairs: electrons and muons?

yes

Table 2: please clarify which fraction. "Transverse-> Transverse polarization (xxx%)".

Fig.2, caption: - L3: -> resultS - L3: "likelihood fit". Please clarify fit of what is done. - L5: "upper limit ... in this analysis". The choice of the normalization is weird. The signal is superimposed after the results are obtained? Also, this was not explained in the text when the Fig. was first introduced.

Include a forward reference in the text to fitting now.

L237: remove "The" at the beginning of the sentence.

L238: remove "The" at the beginning of the sentence.

done

Table 3: does it include electrons and muons? please clarify in the caption, and in the text (L248?).

Clarified in the text.

Fig.3, caption: - "error bars...omitted". That is a weird statement. Do not think it is necessary. Is this a CMS standard?

Removed statement in brackets.

- L3: -> luminosity - L6: please remove "with this analysis". It is unnecessary.

done

L244: sentence does not read well -> final state to be

reworded

L247: suggest removing "parameter".

reworded

L248: suggest removing "four possible". It is unnecessary. L248: four-lepton: does it refer to electrons and muons? Please clarify.

reworded

L251: nS and mS are introduced without being defined.

Named in the next sentence

L252: only nS is referred to. Suggest dropping nS/mS that is unnecessary.

L258: target luminosities: Please define them as it was done earlier.

L265: please remove "clean".

done

L265 and L268: YY was used throughout the text. Please be consistent.

LL270-271: "...so that... models.". It is not clear. Please rephrase.

several of them .. scenarios.

L272: "earlier SM predictions... 10^-5[22]." last sentence is confusing. It really belongs to the introduction.

removed sentence

ARC comments from Jonatan

I think the Abstract is a bit confusing. I would try to make it more clear, in particular adding the data used in the two mentioned recent CMS results.

Reword the abstract

Line 15. Q is not in the right font / format.

Line 19. This description should go in Line 14.

Line 21. Here you use Figure 1, but in Line 19 it's written as Fig. 1. Please use one convention; personally I prefer Figure and Table better than Fig. and Tab.

Line 22. Z-boson --> Z boson

Line 29. BSM theories. [14, 19] --> BSM theories [14, 19]

Line 31. I would group Zrho, Zphi and ZJpsi, stating that the 95% CL limits are two orders of magnitude larger than the SM expectation.

Line 38. branching fraction has appeared before (at least in Line 31) so the symbol should also appear before.

Line 39. at the level of several 10^-5 --> at the 10^-5 level

Line 40. These two phenomenological approaches yield very different results: 10^-5 vs. 10^-9. What are the associated errors?

They assume different importance of direct and indirect amplitudes.

Line 44. How does the precise knowledge of these final states benefit H -> ZZ*?

Strategy section. If space is not a problem, I would try to make this section a bit longer / more easily readable.

Line 56. I'm a bit lost reading at the beginning of the strategy about a slope parameter, when the parameterization hasn't yet been described.

Line 59. New selection criteria with respect to what analysis or paper?

Line 60. What are scaled event sets? Scaled to the HL-LHC lumi?

described below

Line 64. Maybe it would be good to add here that the pileup increase has been also considered.

Include sentence about pileup.

Line 65. luminosity --> luminosities

Line 93. |eta| < 3

Line 132. Define lambda_theta.

Table 1. It doesn't read well when a section starts with a table.

Line 140. I think this line would go better after the different variables (such as dxy or isolation) haven been described.

Line 141. How is the matching chosen between outside-in and inside-out?

Line 148. Is "non prompt" CMS jargon?

Line 152. Why isolation is only requested on the leading muon?

Line 163. Maybe you can state explicitly the Z J/psi threshold.

Line 168. How has this value (2.3, which seems large to me) been determined?

This criterion maintains high efficiency while reducing background, resulting in the best upper limit in combination with the other criteria.

Line 172. Rewrite to something like: The four-muon invariant mass distribution can be seen in Figure 3 for Run-2 (left) and for a representative simulated sample (right).

Line 193. function --> functions

Line 204. Isn't this value too small? Does it have a reference? I'm familiar with luminosity uncertainties in the 2-3% range.

Will add reference. This is a HL-LHC projection (see comment Sezen above)

Line 210. Add "respectively" at the end.

Line 220. PDF already defined in Line 117.

Figures 2 (right) and 3 (right). Shouldn't the title say "CMS simulation" instead of "CMS preliminary"?

ARC comments from Emmanuel

General: - the PAS needs some improvements, in particular Section 2 (Strategy) should better explain what has been done (e.g. the "scaled datasets", 4 mu versus the other channels). Reworded this section

- I would have expected to find answers to some of my questions in the AN, but the AN that is in CADI (v2) contains nothing more than the PAS, basically.

As this analysis is based on the previous Run-2 analyses, many details are found in the AN-2019-232 and AN-2018-106.

Questions to the analysis:

You say you have re-optimised the selection criteria compared to those of the Run 2 paper (HIG-20-008). I understand that, for this optimisation, you simply:

- use the Phase-2 ( ? typo in l 125 ? ) simulated signal

We use Phase-1 simulated samples. The Higgs signal in the two channels under investigation are weakly dependent on mass resolution and scale. The efficiency for the four-lepton final states are expected to be as (very) high as for the present detector with small systematic uncertainties. The detector acceptance though increases, and the effect is estimated at generator level.

- determine the background by a fit to the Run 2 data and scale it up to account for the larger luminosity, center of mass energy, and acceptances.

Correct - after re-optimization.

For the Z J/Psi -> 4 mu channel, which are the re-optimised cuts that reduce the number of selected events in Run 2 from 230 (HIG-20-008) to 51 ( l 161 ) ? i.e., in the slide 7 of the pre-approval talk, how do you go from the top to the bottom plot ? From Tab 1 and the text, and from a quick look into the Run 2 paper, I could not find any difference. Still, there must be one or several cut(s) that have been considerably tightened.

The most important difference are the minimum pT cutoffs. All criteria are described in the text.

The scale-up of the background : how do you determine the increase due to the larger acceptance and improved lepton ID ? The only mention that I found in the PAS is l 123-124, but this holds for the signal. It is not obvious that the background will scale in the same way. (moreover, one would not expect to find this background scaling factor in the paragraph about the simulated signal).

We include more explicit description to distinguish the steps for the background.

Also, the shape of the background at HL-LHC is not necessarily the same as you see it in Run 2, because of the different acceptance. The same holds for the larger center of mass energy - I understand from the pre-approval talk that you use the increase of the signal cross-section (not said in the PAS I think). Which checks have been made in order to make sure that the assumptions that you make here are reasonably valid ?

The background shift in kinematics and increase in acceptance are estimated with generator level simulations. The background distributions are expected to be comparable, assuming trigger and reco efficiencies are still very high.

l 123-124 : if the relative gain in acceptance is 13% in 4mu, and 18% in 2e2mu, i.e. if the gain is more important for electrons than for muons, don't we naively expect more than 18% in 4e ?

This is due to the increase in pseudorapidity, which is larger for e than mu, and dominant for two out of the four leptons.

Tab 1: the mass window 80-100 GeV used to select the Z looks quite loose compared with the resolution that one can gauge from Fig 2, no ?

Combined with the 2lepton and 4lepton vertexing, the candidates include small background but the signal has a tail towards lower mass.

You use some delta-eta, delta-phi and eta(4mu) cuts in the YY channel (l 168-170). Could similar cuts be useful for the Z J/Pi channel too ?

We did not find improvement in the ZJ/psi channel.

Systematic uncertainties, p 6 : (ii) do you take these numbers from the Run 2 analysis ? at HL-LHC, the data-driven uncertainties (like T&P) should be smaller, thanks to samples of much larger statistics. (v) l 221 mentions the Z boson, I guess for its cross-section, but where should sigma(Z) enter here, since your background determination does not use it ? (vi) what is meant by "change in detector conditions" and how this 1(2)% was obtained ?

All systematic uncertainties were found to be very small in the Run-2 samples, already. From Run-2 experience, we include conservatively an uncertainty for not yet quantified variations. Nevertheless, we can drop it as we more likely would account for this by fitting samples by year in the likelihood fit.

l 246-247 : in Run 2, you saw that the background has the same shape in the 3 channels. But this may not hold anymore in HL-LHC. For example, the selected electrons will be more forward than in Run 2, and this can affect the M(4e) distribution.

The acceptance is estimated at generator level for the full phase space. The relative difference in the coverage for the lowest pT leptons does not introduce significant differences in the shape of the four-lepton distributions. We assume that the efficiency does not change significantly either in this interval. As we discussed, we can make these statements only for the two Z-decay mode channels, not the one with J/psi decaying into electrons. We change this.

l 50 : The YY channel is said to be background free. But if you have a sizable rate in H -> YY, you should see H -> Q Z too, and the Z can be off-shell, leading to di-muons in the Y mass region. Do you have any idea of the respective BRs ?

For now. Do you refer to H->ZY with virtual Z under the Upsilon which would result in a non-resonant recoil dilepton? The BF can be expected rare as well, and the di-lepton mass interval is very small. We do not have predictions.

Motivations: You motivate this analysis by the study of Yukawa couplings to quarks (l 12-13) but actually, in the SM at least, these channels are not really sensitive to Hcc, since the indirect amplitudes dominate (middle and right diagrams in Fig 1). By how much should the Hcc coupling be enhanced w.r.t. its SM value, for the H -> Z J/Psi decay to be dominated by the direct amplitude ? and with such an enhancement, wouldn't we be able to measure directly H -> cc ? For H -> YY, is it also the indirect amplitudes that dominate in the SM ? Hbb will be measured directly with a relative good precision at HL-LHC.

The rare decays, at BF 10-5 of lower in the SM, are sensitive to new contributions orders of magnitude smaller than inclusive measurements. Even though models predict up to three orders of magnitude enhancements. In the YY channel, the prediction vary with different assumptions about the importance of direct versus indirect amplitudes. The introduction has been modified.

Some comments on the PAS

l 15 vs 19 : you first use the notation Q for vector mesons, but then it means specifically quarkonia

l 16 : measured -> observed

l 29-30: "Several BSM.. fermions" : already said above,

l 31 : this is about rho and phi, while the paragraph is dealing with quarkonia (cf l 18). Misplaced.

l 34-39: what is the current understanding for the very different BR predictions ? Is it that [26] superceedes the very old (1979) estimation of [22], or is this difference still an open question ? By the way, nowhere you give the BR of H -> Z J/Psi.

Section 2 needs to be reshuffled. What is done became really clear to be only at l 187-188. And more details should be given about the extrapolation to HL-LHC.

l 104: "quarkonium channels" -> "the YY channel". (because Z J/Psi is also a "quarkonium channel").

l 107 : unclear what you mean with this "Y specific trigger": If that exists, you do not seem to use it since you demand 3 muons. Rephrase.

l 111: since you give trigger efficiencies for the single muon trigger, one would expect to see them also for the Y + mu trigger.

l 125 : "Phase 1" ?? "Phase 2" instead ?

l 126 : one usually gives the # of overlaid events.

Tab 1, "dimuons" : could add the opposite charge requirement

l 158 : you already said in l 151 that you require 4 muons

l 160 : jargon

What happens when you have more than 4 muons ? do you require the 4-muon object to be "made from" two selected dimuons ? I guess that the selected events should have at least two dimuons and at least one four-muon candidate, and in case an event had two 4-muon candidates, both would be considered, right ?

There are only single candidate events after all criteria are applied. A choice is not needed.

l 221: these 3.2% and 1.7% "for the Higgs boson and Z boson": assuming you mean their production cross-section, this should be said.

l 222: and here you mean "additional uncertainties" on this cross-section ?

Fig 2: Note that the Run 2 luminosity appears only in the plots. It should appear in the text too. Fig 2 right: suggest "pseudo-data" instead of "data", in the legend.

Comments from Emmanuel on PAS Version 4

l 64 : [30] -> {30, 36]. Suggest to add after [30, 36] : "The extrapolation for HL-LHC presented here, for 3000 fb-1 or 4500 fb-1, follows closely the analysis of the full Run 2 dataset reported in Ref. [36], based on an integrated luminosity of 138 fb-1. "

done

l 66 : add after mu+mu- something like "and the signal is searched for as a resonant peak in the distribution of the invariant mass of the four muons, m_{4mu}." Because I think it's useful to say, early on, that m_{4mu} is your discriminating variable (as opposed to the output of some NN for example).

done

l 67 : remove the sentence about Z -> ee. The mention of this decay better comes only at the end of this paragraph.

removed

l 71-72 : remove "and resonance", since the pT cut on the J/Psi and on the Upsilon is actually unchanged with respect to that of the Run 2 analysis ( 5 GeV), right ?

done (we checked though)

add "increase" before "from 13 to 14"

done

Also, the fact that the best upper limit is achieved with a tighter pT cut is not only due to the harder kinematics due to the higher sqrts, but also to the larger statistics due primarily to the much higher luminosity. All in all, this sentence is not clear.

suggest instead: "The harder transverse momentum spectrum of the decay leptons, which is due to the center-of-mass energy increase from 13 TeV to 14 TeV, is accounted for as explained below. "

done

l 83-85 : remove this sentence, such that this paragraph is only about the increase of the cross-sections due to the higher sqrts.

( What you say here about the luminosity applies not only to the background, but also to the signal of course; I suggest to give these numbers at the very beginning (see above). And what you say at the end of this sentence l 83-85 was already said above (and is less precise here, since the acceptance factor also accounts for the harder pT spectrum). One could instead summarize the scaling of the background in the next paragraph, see below).

l 86 : stop the sentence after "following Ref [36]" ( because you have not explained yet what the "signal region" is, and because it's hard to understand what is done here (e.g. how the Run 2 data is used) without the additional information given in [36]. The reference is enough anyway).

done

next sentence: you've not said yet that the distribution you are looking at is that of the invariant mass of the 4-leptons.

suggest: "For the resulting cuts, the shape of the m_{4mu} distribution for the background is obtained ... fit of the Run 2 data."

done

l 89 : "New samples" -> "Background samples for HL-LHC" I would then add e.g. ", the normalisation being obtained by scaling up the Run 2 background by the aforementioned increases in luminosity, in acceptance, and in cross-section."

done (... in luminosity, acceptance, and cross-section)

l 92 : new paragraph for "According to Ref [36] "

done

Section 1 :

l 13 and l 14-15 are inconsistent with each other. I think there is no need to be exhaustive and tell the full story here. You may try something better for the paper, but here, I would suggest to stick to what really matters: l 13 : only mention SECOND-generation QUARKS and stop the sentence after "reached."

done

l 25-27 : You have not said yet that you look at multi-lepton final states. It is not clear that what you write here would also hold, when the Z and /or the JPsi decays hadronically. Suggest to add at the beginning "When the Z and the meson decay leptonically, ..."

done

l 34-36 : These limits on Z rho and Z phi are not really relevant here, suggest to drop. Instead, add a reference to [30] and [36].

done

Section 4:

l 150 : "For the Z Jpsi final state with either Z or both resonances decaying into e+e− the relative increase is 18%" :

since you said that the acceptance is driven by the low pT leptons, which come in general from the J/Psi, the acceptance gain should be the same for 4-mu and for Z(ee)Jpsi(mumu), no ? i.e. the 18% increase corresponds to Z(mumu)Jpsi(ee) and to 4e ? In which case I suggest to not mention this 18%.

We corrected this statement (see above): the increase of the acceptance is dominated by the increase in pseudo-rapidity range, and it is greater for electrons. Usually, two muons or two electrons fall into the relevant eta range. Please see also the table in the follow up slides. We remove the sentence, as only the 4mu samples are genuinely scaled.

Section 5 :

l 191 - 195: remove (Run 2)

done

l 201 - 203: remove (Run 2)

done

l 216 : end of the line, before the last sentence of the paragraph, add e.g.: "The m_{4mu} distribution of an example HL-LHC sample for an integrated luminosity of 3000 fb-1 is shown in Fig. 2."

done

l 218: remove "in Fig. 2" (now clear).

done

l 222-223 : replace the last sentence by e.g. "The m_{4mu} distribution of an example HL-LHC sample for an integrated luminosity of 3000 fb-1 is shown in Fig. 3, with the fit superimposed."

done

Section 7

l 245 : remove "and Z boson", not used here !

done (corrected both these sentences)

Section 8

l 254 : remove the sentence "Samples are... " - already said.

done

Some details :

all implemented, or further clarified

Abstract: line 6 : "are searched" -> "have been searched". idem in line 8.

l 11 : "model Ref [9]" -> "model [9]"

l 16 : is Ref 14 really needed ?

removed these two lines anyways

l 18 : drop ", which are vector mesons". The reader has been at school.

There are other J^PC quarkonium states - here we only consider vectors

l 45 : "at values of about xxx for B() and yyy for B() ."

l 52: [30] -> [30, 36].

l 55 : I would drop gamma rho and gamma phi

l 124-125 : "The dependence .. states on the average ... interactions is negligible" (swap order).

l 131 : "quarkonium channels" -> "the YY channel" (because Z J/Psi is also a "quarkonium channel")

reworded

l 188 : "Events with at least 4 muons are selected" : drop. Already said in l 181-182.

l 267-268 : "and reduces the Run 2 upper limit" : no. The "Run 2 upper limit" stays as it is... rephrase. idem for l 281.

done

Comments from Michele on PAS Version 4

All implemented, or further clarified below

abstract:

L2: no hyphens: -> "beyond the standard model (BSM)"

L3: same as above

L4: it is not necessarily higher. Please change: higher -> altered

Correct, but of our immediate interest. We write: possibly .. higher

L5: "Recently...LHC." This sentence belongs to the Introduction, not to the abstract. In the abstract start: "Decays ... in four lepton final states."

L8: Please remove "in the same data". Is this about projections of Run2?!

body of the text: LL12-16: not sure why this is relevant here. Please clarify.

removed

Around LL52-52: At some point in the introduction, it would be good to clarify the 2e2mu and 4mu final states are being considered.

L77: "applying" Please clarify/quantify the new "acceptance limits" at HL-LHC.

we add: ... as described in Sec. 3.

Sec.5: Here (as well as in Tab.1), there is no mention of electrons, nor how (if) they are used at all. Please clarify.

Electrons are not explicitly used, only in combination of the two channels Z J/psi with Z(mm,ee) J/psi(mm). We clarify this in the strategy section now.

L168: "single muon trigger". Why not using 2-mu triggers? was it tried? What was the improvement in signal acceptance?

The single muon trigger is used and established for the H->ZZ* analysis. The trigger efficiency is about 99%.

Tab.1, caption: -> the identification criteria

criteria for ... the candidate selection.

L215: "tried". Was it used? any success? please clarify. Perhaps change "tried" to "used".

change to used. Outcome described earlier in line 210/211.

Tab.2, caption: polarization is mentioned. Polarization effects should be introduced earlier, and discuss how they can provide additional (if any) sensitivity to the signal.

Move to Sec 4

Fig.2 and Fig.3, caption, L2: -> results ... are

The result of the fit to this sample.

L283: Please remove "Different types of"

L284: please rephrase to: -> "In this document, the projected sensitivities of two processes are explored for the HL-LHC data period: the Z J/Psi ..."

L286: please remind the final states that were studied.

L288: rather than "gain", please use the branching ratio limits as a main reference for the results.

Comments from Jonatan on PAS Version 4

Abstract. Remove "Recently, ".

Abstract. If the data is not explicitly mentioned, replace "studied in the same data" by "also studied".

Line 11. Remove "Ref." before [9] as you don't use Ref. in other places.

Line 31. and and --> and.

Line 34. exceed unity by how much?

These are upper limit BFs exceeding 100%.

Line 79. Twice already in the text you point to genuine [20] or misidentified meson. What is a misidentified meson? Reference 20 is about prompt and non-prompt J/psi mesons.

Title of Ref [20] - now [32] is "Observation and measurements of the production of prompt and non-prompt J/psi mesons in association with a Z boson .." Removed the first instance of this sentence in Sec. 1, which indeed is redundant.

Line 222. Sometimes you write Figure, like here, other times it's Fig. like in line 218. I recommend using always the same convention.

Style: at beginning of sentence Figure (Table) are to be spelled out.

Systematic uncertainty vi. Can you explicitly mention the range of the branching fraction uncertainties?

Ref [ ] references the exact PDG edition we used.

Line 264. Tab 3 --> Tab. 3

done

I think it's not explicitly said anywhere that the full Run-2 data is 138 fb-1.

Now in Sec. 2

fixed inconsistencies in characters Q and Y (use \rm)

Sezen comments on PAS V10

- Introduction: Please state explicitly that you are performing a projection study from Run 2 to HL-LHC, preferably towards the beginning. This seems to be not stated until section 2.

Add sentence at end of first paragraph

- L64: as high as for the --> as high as they are for the

done

- Run-2 --> Run 2, Phase-II --> Phase-2 (CMS conventions)

done

- Section 3: Please use the official description from this twiki: https://twiki.cern.ch/twiki/bin/viewauth/CMS/FTRCommonChapters

Include the standard text. Keep sentence from previous version that describes the ECAL coverage which is missing in this standard text, but critical for our scaling.

- L123-136: Suggest to move to a more visible place, or into a subsection. Readers may skip the detector section, thinking it is standard.

moved to event reconstruction

- L127-128: a meson-specific trigger

done

- L138-151: include 13 TeV somewhere.

in line 140: ... produced at a cm energy of 13 TeV ...

- L167-168: Since you are using Run 2 samples, it is not clear how you implement this trigger. Please clarify.

Moved trigger description here.

- Sec 5: Are all these selections already present in the Run 2 analysis? Were any Phase-2 modifications implemented? Please state explicitly.

Is now described in first and second sentence of this sec

- 3,000 -> 3000, etc. We do not use comma.

repaired

- Section 7: I believe these systematics are exactly the same as for Run 2? If so, please put a sentence explicitly stating that Run 2 systematics are taken, and no modification was considered for Phase 2.

all but lumi - add this to systematics i) for which also a reference is given

- L243: the 95% CL -> 95% CL

done

- Table 3 caption: upper limits on branching ratios

Done. Also: caption expanded (as presented in approval)

- Sec 9: Paragraph is left-aligned. Please fix.

Approval Minutes

- s6: add the SM multiplication factor in the legend.

done

- Michele: Was using the dilepton trigger never considered?

The triggers are very efficient, they would not improve the sensitivity.

- The ARC will discuss quoting the significance.

Add the sentence after consensus at end of Sec 8

ARC review: No issues.

Figures and tables:

- Use "Phase-2"

corrected

- H x ?? wrt SM in the legend

add 126 x B_SM and 0.2 x B_SM

- move trigger text to the selection section

done

- Table 2 put "-" instead of white space for minpT

done