The paper compares different deep learning approaches to modeling the quantum mechanic properties of molecules, and presents a model that incorporates multiple ideas from physics. Some reviewers appreciated multiple aspects of the paper, including: - A novel approach, offering an interesting contrast to GCN approaches - Compelling numerical results in comparison to a standard GCN approach (even if this not extrapolation, it still shows the benefit of the proposed approach over GCNs). - Discussion of extrapolation metrics in the ML for chemistry field At the same time, a few questions were raised, including: - certain imprecisions and over-claims - LCAO being first-order approximation rather than "basic assumption". The proposed approach does not even use spherical harmonics in the LCAO representation. The fact that this "still works well" is bothering as it necessarily leads to a bad approximation of the molecular orbitals, to a point where the added benefit of chemically informed modeling might be almost gone. - worrysome computational complexity and insufficient discussion thereof - unacceptable argument to not use spherical harmonics to make the approach "simpler for an ML person to understand" The reviewers read the rebuttal and thoroughly discussed it. Overall, the impression is positive and our recommendation is to accept the paper. It is our hope that the authors revised the paper accounting for detailed and thorough comments provided by the reviewers. The AC also invites the Authors to mild the insinuations that ML researchers have no competence in the application domain, as being arrogant and often not true.