<p>Recent advances in imitation learning have enabled robots to perform complex manipulation tasks in unstructured environments. However, several learned policies rely on discrete action chunking, which often introduces discontinuities at the chunk boundaries. These discontinuities degrade motion quality and are particularly problematic in dynamic tasks such as throwing or lifting heavy objects, where smooth trajectories are critical for momentum transfer and system stability. In this work, we present a lightweight post-optimization framework for smoothing chunked action sequences. Our method combines three key components: (1) inference-aware chunk scheduling to proactively generate overlapping chunks and avoid pauses from inference delays, (2) linear blending in the overlapping region to reduce abrupt transitions, and (3) jerk-minimizing trajectory optimization constrained within a bounded perturbation space. The proposed method is validated using a position-controlled robotic arm that performs dynamic manipulation tasks. The experimental results demonstrate that our approach significantly reduces vibration and motion jitter, leading to smoother execution and improved mechanical robustness. The proposed approach requires minimal computational resources and is broadly applicable across different robot platforms and learning methods.</p><p>Project page: <a href="https://sites.google.com/view/action-lipo">https://sites.google.com/view/action-lipo</a></p>