The field of Gaussian Splatting is moving towards more efficient and generalizable methods for dynamic scene reconstruction. Recent developments have focused on improving the speed and accuracy of streaming frameworks, such as Instant Gaussian Stream, which achieves fast and generalizable streaming with an average per-frame reconstruction time of 2s+. Other innovative approaches include rate-aware compression frameworks like 4DGC, which reduces storage size while maintaining superior rate-distortion performance. Dual-Hierarchical Optimization methods, such as DHO, have also shown promising results in semantic 4D Gaussian Spatting. Furthermore, novel frameworks like X$^2$-Gaussian enable continuous-time 4D CT reconstruction by integrating dynamic radiative Gaussian splatting with self-supervised respiratory motion learning. Disentangled 4D Gaussian Splatting has also been proposed, which disentangles temporal and spatial deformations, eliminating the reliance on 4D matrix computations. Noteworthy papers include: Instant Gaussian Stream, which achieves fast and generalizable streaming with a significant reduction in per-frame reconstruction time. 4DGC, which outperforms existing methods in rate-distortion performance across multiple datasets. X$^2$-Gaussian, which advances high-fidelity 4D CT reconstruction for dynamic clinical imaging with a state-of-the-art 9.93 dB PSNR gain.