Transgressions of the Euler class and Eisenstein cohomology of \(\mathrm{GL}_N(\text{Z})\) (Q783810)

The authors expose interesting new connections between the geometry and topology of arithmetic groups on one hand, and the arithmetic of modular forms on the other hand. For instance, they prove the existence of explicit homomorphisms \(\mathrm{H}_{N-1}(\Gamma; \mathbb{Z}) \to M_N(Y)\), where \(\Gamma\) is a congruence subgroup in \(\mathrm{GL}_N(\mathbb{Z})\), \(Y\) is a modular curve, and \(M_N(Y)\) is its space of weight \(N\) modular forms with trivial \(\Gamma\)-action. Letting \(U \subset \Gamma\) be a subgroup of maximal rank \((N-1)\) of positive integral units in a totally real number field \(F\) of degree \(N\) over \({\mathbb{Q}}\) (and embedded in \(\Gamma\) via a regular representation), the authors can evaluate these explicit homomorphisms on the image in \(\mathrm{H}_{N-1}(\Gamma;\thinspace {\mathbb{Z}})\) of the fundamental class in \(\mathrm{H}_{N-1}(U; \mathbb{Z})\), and they obtain a modular form of weight \(N\) which is the restriction to the diagonal of a Hecke-Eisenstein modular form of weight \((1,\dots,1)\) for a congruence subgroup of \(\mathrm{PGL}_2(\mathcal{O}_F\)). This remarkable finding comes out of a larger framework which allows the authors to recover deep classical results like the Klingen-Siegel theorem and Damerell's theorem on Hecke \(L\)-functions of imaginary quadratic fields, and promises results over general number fields, which are announced to be made precise in a more technical paper in preparation by the authors in collaboration with Venkatesh.