ProbXiv
Explorersign in
Problem archiveProblem record

Statement

In other words, we ask whether the complex Rχ(n, 0)∗ ICMDol (n,0) is determined by the restriction to any open subset of An , in particular the locus over which the Hitchin fibration is smooth. The result is a refinement of the celebrated Ngô support theorem; see [19, §7]. The main issue is to determine the support Sn of the decomposition theorem for Rχ(n, d)∗ ICM(n,d) and the mysterious local systems Ln (d). The previous theorem shows that the following dichotomy holds: either the subvarieties Sn have a Hodge- theoretic nature as summands of Rχ(n, d)∗ ICM(n,d) , which happens when n is not d-integral; or they are image of the stratum of a canonical Whitney stratification of MDol (n, d), which happens when n is d-integral. This means that we have two extreme cases: the space MDol (n, 0), i.e. d = 0, is the most singular moduli space, but Rχ(n, 0)∗ ICMDol (n,0) |Ared n has no proper sum- mand; while when gcd(n, e) = 1, MDol (n, e) is smooth with the maximal number of summands of Rχ(n, e)∗ ICMDol (n,e) |Ared n . Remarkably, the local systems Ln (e) that appear in these new summands of Rχ(n, e)∗ ICMDol (n,e) can be expressed in terms of the topology of the symplectic singularities Wn . We could informally say that there exists a conservation law between Hodge-theoretic summands of the decomposition theorem and singular strata as the degree d varies. For these reasons we cal In other words, we ask whether the complex Rχ(n, 0)∗ ICMDol (n,0) is determined by the restriction to any open subset of An , in particular the locus over which the Hitchin fibration is smooth. It suggests that IH(MDol (n, 0)) should be regarded as a primitive indecomposable building block of the cohomology of any other Dol- beault moduli space MDol (n, d).

Record

Comments

Nothing has been published against this problem yet, and nothing has been said about it.

Sign in with an institutional address to take part in the discussion. Reading every thread stays open to everyone.

Sign in

Solve with an agent

Open the statement in a chat, with the problem and the ground rules already written into the prompt.

This opens a third-party site. Nothing is posted back to ProbXiv and nothing you write there is recorded here — what a model gives you is an attempt, which a person still has to check.