Search for $\textit{tH}$ production in the $H \to \tau\tau$ decay mode, and a combination with other searches, using $\textit{pp}$ collisions at 13 TeV and 13.6 TeV with the ATLAS detector

Technical Summary: Search for tH production in the $H \to \tau\tau$ decay mode and combination with other searches

Problem and Motivation
The Yukawa coupling of the Higgs boson to the top quark ($y_t$) is critical for understanding the stability of the electroweak vacuum and the shape of the Higgs potential. While the production of a Higgs boson in association with a top-antitop pair ($t\bar{t}H$) has been observed, the production of a Higgs boson with a single top quark ($tH$) remains elusive. The $tH$ process is uniquely sensitive to the sign of the top-quark Yukawa coupling due to destructive interference between diagrams where the Higgs couples to the top quark and those where it couples to the $W$ boson. This makes $tH$ a sensitive probe for potential new physics, such as a $CP$-odd mixture in the coupling. Previous searches by the ATLAS Collaboration in the $H \to \gamma\gamma$, $H \to b\bar{b}$, and multi-lepton final states have yielded signal strengths compatible with the Standard Model (SM) but with limited significance. This paper addresses the need to improve sensitivity by targeting the $H \to \tau\tau$ decay mode, specifically in final states containing hadronically decaying $\tau$-leptons ($\tau_{\text{had}}$), and combining these results with previous ATLAS measurements.

Methodology
The analysis utilizes proton-proton collision data collected by the ATLAS detector at a center-of-mass energy of $\sqrt{s} = 13$ TeV, corresponding to an integrated luminosity of 140 fb$^{-1}$ (Run 2). The search focuses on the $tHq$ and $tWH$ production modes, where the Higgs boson decays into a $\tau$-lepton pair, with at least one $\tau$ decaying hadronically.

• Event Selection: Two distinct channels are defined based on the final state topology:
  1. $2\ell + 1\tau_{\text{had}}$: Events with two tight light leptons (electrons or muons) and one tight $\tau_{\text{had}}$. The signal region (SR) requires the two light leptons to have the same electric charge (SS), which significantly suppresses reducible backgrounds.
  2. $1\ell + 2\tau_{\text{had}}$: Events with one tight light lepton and two tight $\tau_{\text{had}}$ of opposite charge.
  • Pre-selection criteria include requirements on lepton $p_T$, isolation, and the presence of at least two jets, with one or two identified as $b$-jets.
• Reconstruction:
  • Top Quark: Reconstructed using the leading $b$-jet and the light lepton. The neutrino from the top decay is estimated based on kinematic correlations with the charged lepton.
  • Higgs Boson: The $\tau\tau$ system mass is reconstructed using the Missing Mass Calculator (MMC) method, which utilizes the visible $\tau$ decay products and the missing transverse momentum attributed to the Higgs system.
• Multivariate Analysis: To enhance the signal-to-background ratio, artificial neural networks (NNs) are trained separately for each channel. The NNs distinguish between $tH$ signal events and background processes (dominated by $t\bar{t}$, $t\bar{t}X$, $W/Z$+jets, and dibosons). Input features include kinematic variables of the final state objects, the spectator jet (characteristic of $tHq$), and reconstructed masses.
• Background Estimation: A data-driven approach is employed to estimate backgrounds from misidentified light leptons and fake $\tau_{\text{had}}$ (jets misidentified as $\tau$). Normalization factors (NFs) are derived in dedicated control regions (CRs) enriched in specific background types (e.g., heavy-flavour decays, $\tau$ fakes in different $p_T$ bins) and applied to the signal regions.
• Statistical Combination: The results from the $H \to \tau\tau$ channels are combined with previous ATLAS searches in the $H \to \gamma\gamma$, $H \to b\bar{b}$, and multi-lepton channels. The combination accounts for correlated systematic uncertainties (experimental and theoretical) and uses a profile likelihood fit to extract the signal strength ($\mu_{tH}$).

Key Contributions

• New Channel: This work presents the first ATLAS search for $tH$ production in the $H \to \tau\tau$ decay mode with hadronic $\tau$ decays, utilizing the full Run 2 dataset.
• Improved Background Modeling: The analysis implements a detailed data-driven estimation of $\tau_{\text{had}}$ fakes and misidentified light leptons, utilizing multiple control regions to constrain normalization factors across different kinematic regimes.
• Comprehensive Combination: The paper performs the most comprehensive combination of $tH$ searches to date, integrating the new $H \to \tau\tau$ results with previous ATLAS measurements in $H \to \gamma\gamma$, $H \to b\bar{b}$, and multi-lepton final states. This includes updates to the $H \to b\bar{b}$ analysis to correct for fake-lepton contributions.

Results

• Individual Channel ($H \to \tau\tau$): No significant excess over the expected background is observed. The measured signal strength is $\mu_{tH} = -0.9^{+4.4}_{-4.2}$. An observed (expected) upper limit at the 95% confidence level (CL) on the signal strength is set at 8.6 (9.3) times the SM prediction.
• Combined Measurement: The combination of the $H \to \tau\tau$ results with previous ATLAS searches yields a measured signal strength of:
  $$\mu_{tH} = 3.3^{+1.7}_{-1.5} = 3.3 \pm 1.2 \text{ (stat.)} ^{+1.1}_{-1.0} \text{ (sys.)}$$
  This corresponds to an observed (expected) significance of 2.3 (0.8) standard deviations.
• Combined Limits: The combined observed (expected) 95% CL upper limit on the signal strength is 6.1 (2.7) times the SM prediction.
• Consistency: The measured signal strength is compatible with the SM prediction at the 12% level. When allowing the $t\bar{t}H$ signal strength to float, the results remain largely unchanged, confirming the robustness of the $tH$ measurement against $t\bar{t}H$ contributions.

Significance
The paper claims that this combination represents the most precise measurement of $tH$ production to date. By integrating the new $H \to \tau\tau$ channel with existing searches, the analysis achieves a significantly improved precision compared to previous Run 2 combinations or individual Run 3 measurements. The result provides the strongest constraints to date on the $tH$ production cross-section and the top-quark Yukawa coupling sign within the ATLAS framework, setting the stage for future precision measurements in Run 3 and beyond.