The linearity conjecture for support loci of Ext modules

Let cMcM be a regular holonomic cDXcD_X-module, and let WXW\to X be a relative compact open subset. Suppose that, locally on WW,

cExtcDX,Rl(cNk,cDX,R)0cExt^l_{cD_{X,R}}(cN_k,cD_{X,R})\not=0

for some nln+rn\le l\le n+r. For a support locus Sj(ExtDX,Rl(Nk,DX,R))S_j(\mathscr Ext^l_{\mathscr D_{X,R}}(\mathscr N_k,\mathscr D_{X,R})), the linearity conjecture. If it is nonempty, then it is a finite union of translated (jn)(j-n)-codimensional linear subspaces of Cr\mathbb C^r. If M\mathcal M underlies a Q\mathbb Q-mixed Hodge module, for example M=OX\mathcal M=\mathscr O_X, then all these support loci are defined over Q\mathbb Q. The conjecture is motivated by the relative codimension filtration and known results on regular holonomic maximal extensions; the source notes that it holds for l=nl=n, while the cases with n<ln+rn<l\le n+r remain conjectural.

Sources & referencesView supporting material

Primary source

Lei Wu, “Riemann-Hilbert correspondence for Alexander complexes”, arXiv:2104.06941 (2026).

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.