Abstract:We express the contribution of certain maximal parabolic Eisenstein series to the spectral side of the Arthur--Selberg trace formula for GL$(n)$ in terms of zeroes of Rankin--Selberg $L$-functions, generalizing previous results for GL(2) and Hecke $L$-functions. As applications, we prove a lower bound for the sum of these zeroes, and a base change relation between the zeroes in the case $n=2$ for cyclic extensions of prime degree.
What problem does this paper attempt to address?
The problem that this paper attempts to solve is to express the contribution of certain maximal parabolic Eisenstein series to the spectral side of the trace formula through the non - invariant Arthur - Selberg trace formula and relate it to the zeros of the Rankin - Selberg L - function. Specifically, the author studies the zero problem of the Rankin - Selberg L - function on \( \text{GL}(n) \) by considering the non - invariant Arthur - Selberg trace formula and the Langlands - Shahidi method. The main goal of the paper is to derive a specific expression for the contribution of these zeros to the spectral side of the trace formula, thereby generalizing previous results on \( \text{GL}(2) \) and Hecke L - functions.
### Main Contributions
1. **Theoretical Derivation**: The author derives a specific expression for the contribution of the zeros of the Rankin - Selberg L - function to the spectral side of the trace formula through the method of non - invariant trace formula. This expression involves not only the sum of zeros but also some other distribution terms.
2. **Applications**:
- **Lower Bound of the Sum of Zeros**: The author proves a lower bound estimate for the sum of the zeros of the Rankin - Selberg L - function, which can be regarded as a generalization of the Weil criterion.
- **Base Change Relation**: For the special case of quadratic cyclic extensions, the author derives a base change relation, which relates the zeros of the Hecke L - function to the zeros of the Hecke L - function on the base field.
### Formula Summary
- **Main Theorem**:
\[
a_M^M(s) J_M^{M,P}(f,s)=\sum_{\pi \in \Pi(M(A)^1)} a_G^M \left( \sum_\rho 2\pi h_\pi(\rho)-D_{\pi_1, \pi_2}(f) \right)
\]
where \( \rho \) represents the non - trivial zeros of \( L(s, \pi_1\times\tilde{\pi}_2) \), and \( D_{\pi_1, \pi_2} \) is a specific distribution term.
- **Lower Bound of the Sum of Zeros**:
\[
\sum_\rho h_\pi(\rho)\geq D_{\pi_1, \pi_2}(f)
\]
- **Base Change Relation**:
\[
\sum_{\eta \mapsto \eta_E} \sum_\rho \left( h'_\eta(\rho_\eta)-D_{\mu, \nu}(f') \right)=\sum_\rho \left( \tilde{h}_{\eta_E}(\rho_{\eta_E})-\tilde{D}_{\mu_E, \nu_E}(f) \right)
\]
### Conclusion
Through the above methods, the author not only provides a new way to study the zeros of the Rankin - Selberg L - function but also provides new tools and perspectives for related problems in analytic number theory. In particular, these results have important application values in dealing with low - order zeros and base change relations.