Unlocking Positive Transfer in Incrementally Learning Surgical Instruments: A Self-reflection Hierarchical Prompt Framework

ArXi:2604.02877v1 Announce Type: new To continuously enhance model adaptability in surgical video scene parsing, recent studies incrementally update it to progressively learn to segment an increasing number of surgical instruments over time. However, prior works constantly overlooked the potential of positive forward knowledge transfer, i.e., how past knowledge could help learn new classes, and positive backward knowledge transfer, i.e., how learning new classes could help refine past knowledge.