The 8088 The 8088 ← All news
arXiv cs.CL AI Research Apr 20

SCHK-HTC: Sibling Contrastive Learning with Hierarchical Knowledge-Aware Prompt Tuning for Hierarchical Text Classification

★★★★★ significance 2/5

The paper introduces SCHK-HTC, a new method for few-shot hierarchical text classification that uses sibling contrastive learning and hierarchical knowledge-aware prompt tuning. This approach improves the model's ability to distinguish between semantically similar classes in a label hierarchy. The method outperformed existing state-of-the-art models on several benchmark datasets.

Why it matters Refining few-shot performance in hierarchical structures addresses a persistent bottleneck in deploying specialized LLM-based classification systems.
Read the original at arXiv cs.CL

Tags

#text classification #few-shot learning #contrastive learning #prompt tuning #nlp

Related coverage