The cuspidal cohomology decomposition conjecture for GLrGL_r

Fix a cocompact lattice JZ(\mathbbA)/Z(F)J\subset Z(\mathbbA)/Z(F) and an admissible nn-tuple ω\overline{\omega}. Let

Hcusp,J0(ω)H^0_{\operatorname{cusp},J}(\overline{\omega})

be the cuspidal intersection cohomology representation, and let ρπ,ω\rho_{\pi,\overline{\omega}} denote the representation associated with a cuspidal representation π\pi as in the introduction. Cuspidal cohomology decomposition conjecture. If G=GLrG=GL_r, then, as a representation of GLr(\mathbbA)×ΓF(n)GL_r(\mathbbA)\times\Gamma_{F^{(n)}},

Hcusp,J0(ω)π cuspidal representation of GLr(\mathbbA)π(J)=Id(πρπ,ω).H^0_{\operatorname{cusp},J}(\overline{\omega})\simeq\bigoplus_{\substack{\pi\ \text{cuspidal representation of }GL_r(\mathbbA)\\ \pi(J)=\operatorname{Id}}}\left(\pi\boxtimes\rho_{\pi,\overline{\omega}}\right).

Here ω\overline{\omega} is identified with the corresponding representation of (G^)n(\widehat{G})^n. This is an expected automorphic–Galois decomposition of cuspidal cohomology; the supplied source does not state whether it has been proved or remains open.

Sources & referencesView supporting material

Primary source

Yakov Varshavsky, “Moduli spaces of principal F-bundles”, arXiv:math/0205130 (2004).

Progress summary

Never refreshed

Nothing recorded yet. Refresh searches the literature and the public web for attempts on this problem, and writes the first summary here.

Solutions 0

No solutions have been posted yet.