The adult human brain shows remarkable plasticity following perceptual learning, resulting in improved visual sensitivity. However, such improvements in visual perception commonly require extensive practice and hours of training. Therefore, an ongoing challenge has been to improve the efficiency of such prolonged repetition-based learning. Here we aimed to uncover whether improving the efficiency of repetition-based learning with brief memory reactivations engages distinct neural mechanisms. To address this question, human participants encoded a visual discrimination task. Then, three sessions of brief memory reactivations of only five trials (Reactivation group), instead of hundreds (Repetition group), were conducted on separate days. The Reactivation group exhibited significant learning (mean learning gains = 24.7% ± 4.1% SE), comparable to the Repetition group (25.4% ± 3.9%, p=.90, BF01=3.22), indicating remarkable efficiency by reducing training duration. Following reactivation-induced learning, activity relative to baseline in the bilateral intra-parietal sulcus (IPS) was greater compared to repetition-based learning. Furthermore, resting-state functional connectivity between the inferior parietal lobule and the middle and inferior temporal gyri was reduced following repetition-based learning, but not reactivations, a reduction that was correlated with behavioral improvement. These results suggest that improving the efficiency of repetition-based learning with memory reactivations recruits distinct neural processes while leading to similar behavioral gains, demonstrating enhanced engagement of higher-order control and attentional resources. Temporal-parietal resting functional connectivity changes also imply differential offline perceptual learning and memory representations. These findings shed light on unique mechanisms underlying efficient learning approaches, providing valuable insights for their future applications.