Zero-shot learning (ZSL) aims to recognize unseen classes without visual instances. However, existing methods usually assume clean labels, overlooking real-world label noise and ambiguity, which degrades performance. To bridge this gap, we propose the Dynamic Visual-semantic Alignment (DVSA), a robust ZSL framework for learning from ambiguous labels. DVSA uses a bidirectional visual-semantic alignment module with attention to mutually calibrate visual features and attribute prototypes, and a contrastive optimization grounded in Mutual Information (MI) at the attribute level to strengthen discriminative, semantically consistent attributes. In addition, a dynamic label disambiguation mechanism iteratively corrects noisy supervision while preserving semantic consistency, narrowing the instance-label gap, and improving generalization. Extensive experiments on standard benchmarks verify that DVSA achieves stronger performance under ambiguous supervision.
@article{arxiv.2604.17710,
title = {Dynamic Visual-semantic Alignment for Zero-shot Learning with Ambiguous Labels},
author = {Jiangnan Li and Linqing Huang and Xiaowen Yan and Min Gan and Wenpeng Lu and Jinfu Fan},
journal= {arXiv preprint arXiv:2604.17710},
year = {2026}
}
Comments
Accepted by ICME 2026 (IEEE International Conference on Multimedia and Expo)