This paper proposes an advanced architecture that mitigates difficulties of recurrent matrix multiplications by decomposing A-multiplications into numerous groups and optimizing positional encoding by way of Grouped Finite Impulse Response (FIR) filtering, and incorporates an identical mechanism to enhance the stability and overall performance from