We sought to establish whether torque pulses applied by an exoskeleton to the hip and knee joint modulate propulsion mechanics and whether changes in propulsion mechanics would be sustained after exposure to torque pulses under user-driven treadmill control. We applied twelve different formulations of torque pulses consecutively over 300 strides to 24 healthy participants, and quantified the evolution of four outcome measures – gait speed (GS), hip extension (HE), trailing limb angle (TLA), normalized propulsive impulse (NPI) – before, during, and immediately after training. We tested whether the pulse conditions modulated propulsion mechanics during and after training relative to baseline. Metrics of propulsion mechanics significantly changed both during and after training. After training, HE, NPI, and GS significantly increased in eleven conditions, three conditions, and four conditions, respectively. Increases in HE during and after training were observed in conjunction with hip/knee flexion pulses during early stance, or hip/knee extension during late stance. Increases in NPI during training were associated with hip/knee extension during early stance, or knee flexion during late stance. Knee flexion during early stance resulted in positive after-effects in NPI. Increases in GS were associated with the application of hip flexion pulses. Conditions exhibiting the largest positive changes in HE, and not NPI, during training resulted in increased GS after training. Analysis of the relationship between the effects measured during and after training suggests that, when present, after-effects arise from retention of training effects, and that retention is specific to the component of propulsion mechanics affected by training.