The Impact of Semantic Pairs on Self-Supervised Representation Learning

ArXi:2510.08722v3 Announce Type: replace-cross Instance discrimination learns visual representations by treating different augmented views of the same image as positive pairs. While this encourages invariance to handcrafted transformations, same-image positives can preserve nuisance correlations such as background, texture, illumination, and object-specific details. Semantic positive pairs, i.e., different same-class instances, may reduce these correlations by presenting objects across diverse contexts.