1: \begin{abstract}

2: In this work, the infinite GMRES algorithm, recently proposed by Correnty

3: \emph{et al.}, is employed in contour integral-based nonlinear eigensolvers,

4: avoiding the computation of costly factorizations at each quadrature node to

5: solve the linear systems efficiently.

6: Several techniques are applied to make the infinite GMRES memory-friendly,

7: computationally efficient, and numerically stable in practice.

8: More specifically, we analyze the relationship between polynomial eigenvalue

9: problems and their scaled linearizations, and provide a novel weighting

10: strategy which can significantly accelerate the convergence of infinite GMRES

11: in this particular context.

12: We also adopt the technique of TOAR to infinite GMRES to reduce the memory

13: footprint.

14: Theoretical analysis and numerical experiments are provided to illustrate the

15: efficiency of the proposed algorithm.

16: \end{abstract}

17: