Avoidance for set-theoretic solutions of mean-curvature-type flows (Q6078189)

I think that the best review of this paper is to extract some excerpts from the Introduction to the paper. ``Under mean curvature flow, an initially smooth compact hypersurface in \(\mathbb{R}^{n+1}\) must become singular in finite time. Singularities typically occur before the surface disappears, that is, before its area tends to zero. Thus it is desirable to have weak notions of mean curvature flow that allow the flow to extend past singularities.'' And a few lines below. ``A key feature of weak set flows is that not only do they not bump into smooth mean curvature flows, they also cannot bump into other weak set flows. More precisely, they satisfy the following avoidance principle: two initially disjoint weak set flows remain disjoint as long as at least one of them remains compact. (Under the mild hypothesis that the ambient space is complete with Ricci curvature bounded below, any initially compact weak set flow remains compact.) Ilmanen gave a very elegant proof of the avoidance principle in Euclidean space, but it strongly relied on invariance of mean curvature flow under spatial translations, and thus it did not seem to extend to other Riemannian manifolds. One of the main contributions of this paper is modifying Ilmanen's proof so that it works in arbitrary Riemannian manifolds, and, more generally, for closed sets (in a Riemannian manifold) moving by mean curvature plus an ambient vectorfield.'' Among the results included in the paper, I consider that the following deserves some attention. ``In the appendix, we give a simple proof of a version of Ilmanen's interpolation theorem, a key tool in the proof of the avoidance theorem.''