1: \begin{abstract}

2:     Accurately simulating non-Markovian quantum dynamics in system-bath coupled

3:     problems remains challenging. In this work, we present a novel memory

4:     truncation scheme for the iterative Quasi-Adiabatic Propagator Path Integral

5:     (iQuAPI) method to improve accuracy. Conventional memory truncation in

6:     iQuAPI discards all influence functional beyond a certain time interval,

7:     which is not effective for problems with a long memory time. Our proposed

8:     scheme selectively retains the most significant parts of the influence

9:     functional using the density matrix renormalization group algorithm. We

10:     validate the effectiveness of our scheme through simulations of the

11:     spin-boson model across various parameter sets, demonstrating faster

12:     convergence and improved accuracy compared to the conventional scheme. Our

13:     findings suggest that the new memory truncation scheme significantly

14:     advances the capabilities of iQuAPI for problems with a long memory time.

15: \end{abstract}

16: