Contrasting ultrafast light-driven electron-hole interaction dynamics in monolayer MoS₂ and metallic NbSe₂

Referee: Methods section (TD-HSEX implementation): The central contrast between MoS2 and NbSe2 rests on the assumption that the screened-exchange kernel adequately captures electron-hole interactions in the metallic case. In NbSe2 the Fermi level lies inside bands, so intraband screening and possible frequency dependence during the sub-cycle drive may require corrections beyond the static or adiabatic screening employed; if these alter the reported yield enhancement or phase shifts by amounts comparable to the MoS2-NbSe2 difference, both the material contrast and the Keldysh interpretation weaken. A sensitivity test or benchmark against frequency-dependent screening would be needed.

Authors: We thank the referee for pointing out this important caveat for the metallic case. Our implementation employs a static, momentum-dependent screened-exchange kernel obtained from the 2D RPA dielectric function, which is a standard approximation in multiband density-matrix simulations of TMD monolayers. While a fully dynamical, frequency-dependent treatment would be desirable for capturing sub-cycle screening in metals, the primary material contrasts we report originate from the distinct band dispersions, optical resonances, and the specific bands contributing to the emission process rather than from fine details of the screening. To address the concern, we will add a dedicated paragraph in the revised Methods section discussing the adiabatic approximation and its expected range of validity for NbSe2. We will also include a sensitivity analysis in the supplementary material by varying the effective screening strength and dielectric cutoff, demonstrating that the reported yield ratios, phase shifts, and angular patterns remain qualitatively unchanged. This will support the robustness of the MoS2–NbSe2 contrast and the Keldysh tunneling interpretation. revision: partial

Referee: Results section (HHG spectra and carrier injection): The claims of yield enhancement, phase modification, angular-pattern changes, and qualitative Keldysh agreement are presented without quantitative validation, error bars, convergence checks (k-grid, basis size), or comparison to experimental HHG data or independent calculations for either material. Because the soundness of the TD-HSEX outputs is load-bearing for all interpretive statements, the absence of such benchmarks leaves the reported differences and the proposed framework on uncertain footing.

Authors: We agree that explicit convergence and validation data would increase confidence in the numerical results. In the revised manuscript we will add a new subsection (or supplementary section) presenting convergence tests with respect to k-point sampling density and the number of bands retained in the basis; these tests will show that the HHG yields and phases are stable to within a few percent once the reported grid and basis sizes are reached. We will also provide rough numerical uncertainty estimates derived from these convergence studies. Direct experimental HHG spectra for monolayer NbSe2 under the simulated strong-field conditions are not yet available in the literature, but we will expand the discussion to include qualitative comparisons with existing HHG measurements on MoS2 and other TMD monolayers, noting consistency in the observed harmonic orders and polarization dependence. While quantitative error bars on many-body simulations are inherently approximate, the added convergence data will place the reported material contrasts on firmer numerical ground. revision: partial