AI RESEARCH
HyCal: A Training-Free Prototype Calibration Method for Cross-Discipline Few-Shot Class-Incremental Learning
arXiv CS.CV
•
ArXi:2604.15678v1 Announce Type: new Pretrained Vision-Language Models (VLMs) like CLIP show promise in continual learning, but existing Few-Shot Class-Incremental Learning (FSCIL) methods assume homogeneous domains and balanced data distributions, limiting real-world applicability where data arises from heterogeneous disciplines with imbalanced sample availability and varying visual complexity.