General:
There is excessive use of the term 'significance', which could be reduced. The purpose should always be to measure a physical quantity (e.g. signal strength of EWK WZjj production). The significance is just a tool to quantify an excess that is observed with respect to some null hypothesis. It is not an observable that we can measure.

In addition, the null hypothesis for the different interpretations is not always clear. Especially in a paper where so many different interpretations are given it needs to be very clear what has been done. In the case of the EWK WZjj signal strength measurement, the null hypothesis is 'no EWK WZjj', which is not the same as the null hypothesis for the BSM interpretations (because the null hypothesis for BSM interpretations is 'SM' and that includes EWK WZjj)

type A
line 5: 'to the Higgs' - should this still say 'Higgs boson'?

line 16: 'These interactions include WZ quartic couplings, as shown in Fig. 1 (a)' - this is repetitive as line 14 already state 'quartic WZ 
interactions are accessible through triple vector boson production or through vector boson scattering. Perhaps better to replace 'These interactions .. shown in Fig.1 (a)' with 'This is illustrated in Fig.1 (a)' or similar. 

line 73: spurious quotation mark at the end

line 234: 'a EW signal selection' -> 'an EW signal selection'

Figure 2: If using (a) and (b) subcaptions under the plots, then why not use these instead of (left) and (right) in the caption? Alternatively, drop the (a) and (b) subcaptions. 

line 356: 'Extraction of the significance' -> 'Extraction of the signal strength'

Fig 4: caption: shouldn't this be standalone (ie not refer to the caption of another figure)? In general the captions of these figures are very long, is there any information that can be dropped?

Figure 6: The caption refers to (top) and (bottom) figures when the two plots are actualy side-by-side.


type B:
Abstract: No significant excess is observed, why not highlight the mu_(VBS) best-fit signal strength and uncertainties in the abstract instead? 

line 234-235: 'EW signal selection control region' sounds confusing (as if the signal selection is used to define a control region) when really it is a control region that is as close as possible to the EW signal selection, right? This could be clarified. In addition, it would probably  make sense to mention here already that this control region is included in the fit? 

line 291-294: "Higher order EW corrections in VBS processes are known to be negative and at the level of tens
of percent [51]. Such corrections have not been calculated for the final state considered in this
 paper, and therefore are not explicitly considered. It is, however, expected that the contribution
of such corrections are within the theoretical uncertainties considered." - Would the correction really have an effect of 10s of percent? How large is the theoretical uncertainty assigned to try to cover this?

Table 2: how is the relative systematic uncertainty calculated? is it (nominal^2 - (uncert with set of NP's frozen)^2)/(nominal^2); sqrt(nominal^2-(uncert with set of NP's frozen^2))/nominal; or yet something else? On the effect on the significance: is the nuisance term actually removed from the fit, or is it frozen to its best-fit value? If it really is the former then the best-fit mu value can also change (if there are pulls on these parameters) and so this is not a fair comparison. It would be more insightful to use the relative uncertainty on mu_EWK calculated in the same fashion as for the fiducial WZjj cross section measurement. 

line 321 'uncertainty in the ... EW WZ significance measurement': there is no such thing as the measurement of a significance; the measurement is of the EW WZ signal strength, and after the measurement one can quantify the significance of the observed excess over the background-only expectation (hypothesis: no EW WZ production). Please rephrase

line 324-325 "The cross section for WZjj production is measured with a combined maximum likelihood fit to the observed and expected event yields". Rephrase:  The fit is not to the observed and expected event yields, but of the expected event yields to the observed event yields (assuming the four individual decay channels indeed include a single bin in the fit each).

line 353-355: "This motivates the use of ...for the extraction of the EW WZjj significance". Rephrase: for the measurement of the EW WZjj signal strength.

line 357: Is a distribution fitted in this control region too, or is the control region included as a counting experiment?

line 358: "The fit is performed independently for each channel" - if there are independent fits for each channel then for each channel a different muEWK should be quoted. Only one is given. So it looks more like a simultaneous fit of all the different channels. If there genuinely is one fit per channel, then the authors need to explain how they arrive at a single number at the end.

line 360-361: 1) It is not clear from this statement what those pdfs are. 2) what are the correlations between different sources of uncertainty? In CMS the correlation between different uncertainty sources, a priori,  is usually either -100%, 0%, or 100%. 

between line 365 and 366: re-order the significance and best-fit signal strength values. 

General in section 9: it is not fully clear what is actually happening here. Is it correct that the aQGC parameters transform the EWK WZjj shape and norm? In that case the EWK WZjj contribution shown in Figure 4 is not strictly a background process; it is the null hypothesis and any other hypothesis (ie any of the couplings nonzero) does not contain exactly the EWK WZjj contribution as it is shown in Fig 4, is that correct?

line 390: There is not a single fit in the CLs method, but many. Would remove 'extracted from a maximum likelihood fit' and just leave 'extracted using the CLs method'. Additional comment: setting all parameters except for the coupling being probed to 0 is a valid assumption, but did you also try this leaving the unprobed parameters floating?

line 427: "extract the significance of EW WZ production" -> measure the signal strength of EW WZ production. Also quote the value actually measured here instead of only the significance