The integral basic-element conjecture for affine Deligne–Lusztig varieties

At least 2 years old · documented by

Let x∈W~x\in\widetilde W and let [b]∈B(G)[b]\in B(G) be integral, meaning that defect⁡(b)=0\operatorname{defect}(b)=0. For each v∈LP⁡(x)v\in\operatorname{LP}(x) and u∈Wu\in W, define the multiset

E(u,v)={e∣(ω,e)∈wts⁡(u⇒σ(wu)⇢σ(wv)), u−1μ−ω=λ(b)∈X∗(T)Γ}m.E(u,v)=\{e\mid(\omega,e)\in\operatorname{wts}(u\Rightarrow\sigma(wu)\dashrightarrow\sigma(wv)),\ u^{-1}\mu-\omega=\lambda(b)\in X_\ast(T)_\Gamma\}_m.

Set max⁡∅=−∞\max\emptyset=-\infty, and define

d=max⁡u∈Wmin⁡v∈LP⁡(x)max⁡(E(u,v))∈Z≥0∪{−∞},d=\max_{u\in W}\min_{v\in\operatorname{LP}(x)}\max(E(u,v))\in\mathbb Z_{\geq0}\cup\{-\infty\}, c=∑u∈Wmin⁡v∈W(multiplicity of d in E(u,v))∈Z≥0.c=\sum_{u\in W}\min_{v\in W}\bigl(\text{multiplicity of }d\text{ in }E(u,v)\bigr)\in\mathbb Z_{\geq0}.

Let DD and CC be defined by the dimension formula and component-orbit count above. The integral basic-element conjecture. The predictions are: (a) if for every u∈Wu\in W there is some v∈LP⁡(x)v\in\operatorname{LP}(x) with E(u,v)=∅E(u,v)=\emptyset, equivalently d=−∞d=-\infty, then Xx(b)=∅X_x(b)=\emptyset; (b) if Xx(b)≠∅X_x(b)\neq\emptyset, then D≤dD\leq d; (c) if Xx(b)≠∅X_x(b)\neq\emptyset and D=dD=d, then C≤cC\leq c; (d) if [b][b] satisfies ⟨ν(b),α⟩≥1\langle\nu(b),\alpha\rangle\geq1 for every α∈Φ+\alpha\in\Phi^+, then Xx(b)≠∅X_x(b)\neq\emptyset and D=dD=d.

References

Primary source

Felix Schremmer, “Affine Deligne-Lusztig varieties via the double Bruhat graph II: Iwahori-Hecke algebra”, arXiv:2306.06873 (2024).

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.