The goal of this Polymath project is to give a new approach to the Russian proofs of Lusztig's conjectures using higher representation theory.

(A) Kazhdan-Lusztig conjecture (proven by Beilinson-Bernstein).

(A.1) Using Bernstein-Frenkel-Khovanov and generalizations (Sussan, Stroppel-Mazorchuk, etc), the Kazhdan-Lusztig conjectures in type A are equivalent to the following statement: the classes of the simple modules in their categorification correspond to a "dual canonical basis" in a tensor product representation of $\mathfrak{sl}_k$ (for appropriately chosen $k$).

(A.2) Using Losev-Webster and Webster, that statement can be deduced. They construct a theory of tensor product categorifications in type A (i.e. existence + uniqueness).

(Q.1) Can [Losev-Webster] be simplified, so that the connection to Soergel's J.AMS paper (www.ams.org/jams/1990-03-02/S0894-0347-1990-1029692-5/) becomes clear? The goal is to construct a theory, with the input data being a Dynkin diagram (i.e. a simple Lie algebra).

(Q.2) In types B/C/D, the correct set-up is "quantum symmetric pairs", following the paper Bao-Shan-Wang-Webster. What obstacles does one encounter when constructing an analogue of [Losev-Webster] for $\mathfrak{g}=\mathfrak{sp}_4$? Rouquier's theory of tensor product categorifications is also relevant (and hasn't been published yet).