The field of surgical video analysis is witnessing significant advancements with the development of innovative methods for point tracking, disparity estimation, and surgical scene segmentation. Researchers are exploring semi-supervised learning approaches to adapt models to real-world surgical videos, addressing challenges such as domain shift and scarcity of labeled data. Noteworthy papers in this area include SurgTracker, which presents a semi-supervised framework for adapting synthetic-trained point trackers to surgical video, and DGORNet, which refines disparity maps by leveraging monocular depth information. Another notable work is ReSurgSAM2, a two-stage surgical referring segmentation framework that achieves substantial improvements in accuracy and efficiency. MrTrack, an aspiration needle tracker with a mamba-based register mechanism, also demonstrates superior performance in tracking needles with rapid reciprocating motion. These developments highlight the potential for improved surgical quality and patient outcomes through enhanced computer-assisted surgery.