Refined K3 Vafa–Witten generating-series conjecture

Let SS be a K3 surface, let rr be a positive integer, and let VWr,n(t)\mathsf{VW}_{r,n}(t) denote the refined Vafa–Witten invariant for charge (r,0,n)(r,0,n). Let [d]t=(td/2td/2)/(t1/2t1/2)[d]_t=(t^{d/2}-t^{-d/2})/(t^{1/2}-t^{-1/2}), and define

Δ~(q,t)=qk=1(1qk)20(1tqk)2(1t1qk)2.\widetilde\Delta(q,t)=q\prod_{k=1}^{\infty}(1-q^k)^{20}(1-tq^k)^2(1-t^{-1}q^k)^2.

Assume that OS(1)\mathcal O_S(1) is generic. Refined K3 generating-series conjecture.

nVWr,n(t)qn=dr1[d]t2drqrj=0r/d1Δ~(e2djπi/rqd2/r,td)1.\sum_n\mathsf{VW}_{r,n}(t)q^n=\sum_{d\mid r}\frac{1}{[d]_t^2}\frac{d}{r}q^r\sum_{j=0}^{r/d-1}\widetilde\Delta\left(e^{2dj\pi i/r}q^{d^2/r},t^d\right)^{-1}.

This is the generating-series reformulation of the refined K3 multiple-cover conjecture; the general identity remains conjectural whenever it is not covered by the proved prime case.

Sources & referencesView supporting material

Primary source

Richard P. Thomas, “Equivariant K-theory and refined Vafa-Witten invariants”, arXiv:1810.00078 (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.