English

Sub-convexity problem for Rankin-Selberg $L$-functions

Number Theory 2018-07-31 v1

Abstract

We establish a sub-convexity estimate for Rankin-Selberg LL-functions in the combined level aspect, using the circle method. If pp and qq are distinct prime numbers, ff and gg are non-exceptional newforms (modular or Maass) for the congruence subgroups Γ0(p)\Gamma_0(p) and Γ0(q)\Gamma_0(q) (resp) with trivial nebentypus, then for all ϵ>0\epsilon >0 we show that there exists an A>0A >0 such that L(12+it,f×g)ϵ,μf,μg(1+t)A(pq)1/2+ϵmax{p,q}164. L\left(\frac{1}{2}+it, f \times g \right) \ll_{\epsilon,\mu_f, \mu_g}(1+|t|)^A \frac{(pq)^{1/2+\epsilon}}{\max\{p,q \}^{\frac{1}{64}}}. The dependence on μf\mu_f and μg\mu_g, the parameters at infinity for ff and gg respectively, is polynomial. Further, if pp is fixed and qq \rightarrow \infty, we improve this to L(12+it,f×g)ϵ,μf,μg(p(1+t))Aq1212θ27+28θ+ϵ, L\left(\frac{1}{2}+it, f \times g \right) \ll_{\epsilon,\mu_f,\mu_g}(p(1+|t|))^Aq^{\frac{1}{2}-\frac{1-2\theta}{27+28\theta}+\epsilon} , where θ\theta is the exponent towards Ramanujan-conjecture for cuspidal automorphic forms. Unconditionally, we can take θ=7/64\theta = 7/64. This improves all previously known sub-convexity estimates in this case.

Keywords

Cite

@article{arxiv.1807.11092,
  title  = {Sub-convexity problem for Rankin-Selberg $L$-functions},
  author = {Chandrasekhar Raju},
  journal= {arXiv preprint arXiv:1807.11092},
  year   = {2018}
}
R2 v1 2026-06-23T03:18:19.044Z