Search for Higgs boson production at high transverse momentum in the WW decay channel in proton-proton collisions at $\sqrt{s}$ = 13 TeV

Abstract

A search for Higgs boson (\PH) production at high transverse momentum ($p_\mathrm{T}$) in the WW decay channel is presented. The analysis uses proton-proton collisions at $\sqrt{s}$ = 13 TeV recorded by the CMS experiment in 2016$-$2018, corresponding to an integrated luminosity of 138 fb$^{-1}$. The visible decay products of the Higgs boson are reconstructed as a single large-radius jet with one isolated lepton or none (1$\ell$ and 0$\ell$, respectively; $\ell$ = e, $μ$). The \PH-candidate jets are identified using an advanced transformer-based algorithm and are calibrated with the Lund jet plane reweighting technique. The 1$\ell$ channel is further split into gluon fusion, vector boson fusion, and associated production with hadronically decaying vector boson categories, while the 0$\ell$ channel considers all production processes inclusively. The measured cross section times the H $\to$ WW branching fraction relative to the standard model expectation is $μ$ = $-$0.19$^{+0.48}_{-0.46}$, indicating no evidence of a signal above the background. This measurement represents the first dedicated study of highly Lorentz-boosted H $\to$ WW decays, complementing earlier searches for high-$p_\mathrm{T}$ Higgs boson in other decay channels.

Figures (10)

• Figure 1: Illustration of the event topologies analyzed. Right: boosted Higgs boson final states from the ${ \mathup{{{H}}{} _{ {}} ^{ {}}} }\xspace\to{ \mathup{{{W}}{} _{ {}} ^{ {}}} }\xspace{ \mathup{{{W}}{} _{ {}} ^{ {}}} }\xspace\to{ \mathup{{{\ell}}{} _{ {}} ^{ {}}} }\xspace{ \mathup{{{\upnu}}{} _{ {}} ^{ {}}} }\xspace{ \mathup{{{q}}{} _{ {}} ^{ {}}} }\xspace{ \mathup{{{q}}{} _{ {}} ^{ {}}} }\xspace/{ \mathup{{{q}}{} _{ {}} ^{ {}}} }\xspace { \mathup{{{q}}{} _{ {}} ^{ {}}} }\xspace{ \mathup{{{q}}{} _{ {}} ^{ {}}} }\xspace{ \mathup{{{q}}{} _{ {}} ^{ {}}} }\xspace$ decay. Left: associated jets corresponding to the different production processes. From upper to lower: $0{ \mathup{{{\ell}}{} _{ {}} ^{ {}}} }\xspace$ inclusive (all production and decay modes), and the $1{ \mathup{{{\ell}}{} _{ {}} ^{ {}}} }\xspace$ ggF, VBF, and VH production processes.
• Figure 2: Performance curves showing the identification probability of background jets versus ${ \mathup{{{H}}{} _{ {}} ^{ {}}} }\xspace \to { \mathup{{{W}}{} _{ {}} ^{ {}}} }\xspace { \mathup{{{W}}{} _{ {}} ^{ {}}} }\xspace$ signal jets for ParT and ParT-Finetuned. Left: Discrimination performance of the ParT model for various ${ \mathup{{{H}}{} _{ {}} ^{ {}}} }\xspace \to { \mathup{{{W}}{} _{ {}} ^{ {}}} }\xspace { \mathup{{{W}}{} _{ {}} ^{ {}}} }\xspace$ decays against the dominant QCD multijet background. Right: Comparison of $P({ \mathup{{{H}}{} _{ {}} ^{ {}}} }\xspace_{1{ \mathup{{{\ell}}{} _{ {}} ^{ {}}} }\xspace})$ before and after fine-tuning, following the event selection in the $1{ \mathup{{{\ell}}{} _{ {}} ^{ {}}} }\xspace$ channel. The background includes jets originating from QCD multijet events, ${ \mathup{{{W}}{} _{ {}} ^{ {}}} }\xspace({ \mathup{{{\ell}}{} _{ {}} ^{ {}}} }\xspace{ \mathup{{{\upnu}}{} _{ {}} ^{ {}}} }\xspace)$+jets, and top quark processes.
• Figure 3: The distributions for the total simulated background and total signal (scaled by a factor of 3$\times10^4$ for visibility) passing event selection in the $0{ \mathup{{{\ell}}{} _{ {}} ^{ {}}} }\xspace$ channel. The signal is split into classes as defined in the text. The upper left and upper right panels show the soft-drop mass and ParT score distributions for the $\mathup{{{H}}{} _{ {}} ^{ {}}}$ -candidate jet (j) $P({ \mathup{{{H}}{} _{ {}} ^{ {}}} }\xspace_{0{ \mathup{{{\ell}}{} _{ {}} ^{ {}}} }\xspace})$, respectively. The lower left and lower right panels display the $p_{\mathrm{T}}\xspace^\text{miss}\xspace/p_{\mathrm{T}}\xspace^{\text{j}}\xspace$ ratio and the angle $\lvert \Delta\phi(\text{j}, {\vec{p}}_{\mathrm{T}}^{\mkern3mu\text{miss}}\xspace) \rvert$, respectively. Vertical lines indicate the selection conditions imposed to define the SRs.
• Figure 4: Illustration of the SRs and CRs, and the TFs used to relate the QCD background in the different regions (left). The TFs used to predict the QCD process in the four SRs as a function of the $m^*_{\text{j}}$ (right).
• Figure 5: Post-fit $m^*_{\text{j}}$ distributions in the $0{ \mathup{{{\ell}}{} _{ {}} ^{ {}}} }\xspace$ channel, showing the predicted background with total uncertainty, observed data, and the expected pre-fit signal scaled by the labeled strength $\mu$. From left to right, upper to lower, the plots correspond to $\text{SR}_\text{1a}$, $\text{SR}_\text{2a}$, $\text{SR}_\text{1b}$, and $\text{SR}_\text{2b}$. The lower panel of each plot presents the pull distribution, as well as the $\sigma_\text{fit}$ normalized to the $\sigma_\text{stat}$.