AI RESEARCH
SCHK-HTC: Sibling Contrastive Learning with Hierarchical Knowledge-Aware Prompt Tuning for Hierarchical Text Classification
arXiv CS.CL
•
ArXi:2604.15998v1 Announce Type: new Few-shot Hierarchical Text Classification (few-shot HTC) is a challenging task that involves mapping texts to a predefined tree-structured label hierarchy under data-scarce conditions. While current approaches utilize structural constraints from the label hierarchy to maintain parent-child prediction consistency, they face a critical bottleneck, the difficulty in distinguishing semantically similar sibling classes due to insufficient domain knowledge. We.