Digital compute-in-memory (DCIM) architectures are gaining importance for real-time and accurate deep neural network (DNN) inference due to their capacity for precise computations. However, traditional DCIM systems often face challenges in balancing precise data processing with computational efficiency. In scenarios where exact computations are not always necessary, especially in DNNs characterized by sparse data, existing multi-bit operations frequently fail to strike an ideal balance between accuracy and efficiency. In this paper, we introduce a novel approach to DCIM architecture that selectively skips non-critical computations by using probabilistically determined values, effectively reducing the computational load. Our simulation results demonstrate that this method significantly reduces the number of computing cycles required, achieving a speedup of up to 1.5x compared to traditional methods, while maintaining accuracy with only a minimal decrease of 1.5%.