Semantic Hierarchy-Aware Hyperbolic Representations for Multi-Label Classification With Single Positive Labels
IEEE Signal Processing Letters · 2025 · 32 · 3340-3344
DOI: 10.1109/lsp.2025.3596465 ↗Kurz gefasst
Fachartikel in der Zeitschrift IEEE Signal Processing Letters, 2025. Verfasst von Tongtong Liu und 3 weiteren Autor:innen. Die Arbeit wurde bisher 2-mal zitiert (Crossref) und nennt 35 Literaturangaben. Gefördert von: Science and Technology Development Plan of Jilin Province.
Abstract (Originalsprache)
Single positive multi-label learning (SPML) aims to recognize multiple categories with limited supervision from one positive label in an image. With the emergence of pre-trained visual-language models such as CLIP, recent studies focused on capturing label-to-label dependencies. However, hierarchies with deeper layers of labels or more branches in label-to-label relationships cannot be well expressed in Euclidean space. To address the challenge, we introduce a semantic hierarchy-aware hyperbolic representations framework for single positive multi-label learning. Specifically, drawing inspiration from semantic hierarchical information, we introduce a label relation prior strategy to map single labels to other labels. The semantic chain of labels is extracted along the hierarchical path from the child node to the parent node. Furthermore, hyperbolic entailment constraints are adopted to enforce the semantic similarity between image-text pairs and the hierarchical consistency among labels in hyperbolic space. Experimental results conducted on four SPML benchmark datasets demonstrate that our SHHNet achieves state-of-the-art performance.
Förderung
- Science and Technology Development Plan of Jilin Province
Zitieren
APA
Liu, T., Chen, G., Wang, Y., & Li, W. (2025). Semantic Hierarchy-Aware Hyperbolic Representations for Multi-Label Classification With Single Positive Labels. IEEE Signal Processing Letters, 32, 3340-3344. https://doi.org/10.1109/lsp.2025.3596465
Vancouver
Liu T, Chen G, Wang Y, Li W. Semantic Hierarchy-Aware Hyperbolic Representations for Multi-Label Classification With Single Positive Labels. IEEE Signal Processing Letters. 2025;32:3340-3344. doi:10.1109/lsp.2025.3596465
BibTeX
@article{Liu2025,
author = {Liu, Tongtong and Chen, Guoqiang and Wang, Ying and Li, Wenhui},
title = {Semantic Hierarchy-Aware Hyperbolic Representations for Multi-Label Classification With Single Positive Labels},
journal = {IEEE Signal Processing Letters},
year = {2025},
volume = {32},
pages = {3340--3344},
doi = {10.1109/lsp.2025.3596465}
}
EUVIMED