Evolutionary Generalized Zero-Shot Learning

arXiv:2211.13174v2 Announce Type: replace-cross
Abstract: Attribute-based Zero-Shot Learning (ZSL) has revolutionized the ability of models to recognize new classes not seen during training. However, with the advancement of large-scale models, the expectations have risen. Beyond merely achieving zero-shot generalization, there is a growing demand for universal models that can continually evolve in expert domains using unlabeled data. To address this, we introduce a scaled-down instantiation of this challenge: Evolutionary Generalized Zero-Shot Learning (EGZSL). This setting allows a low-performing zero-shot model …

arxiv cs.cv cs.lg cs.ne generalized replace type zero-shot