Holistic Optimal Label Selection for Robust Prompt Learning under Partial Labels

TL;DR

This paper introduces HopS, a holistic label selection method for prompt learning with partial labels, combining local density filtering and global optimal transport to improve robustness and performance.

Contribution

The paper proposes a novel holistic label selection framework combining local density filtering and global optimal transport for prompt learning under partial labels.

Findings

HopS outperforms all baselines on eight benchmark datasets.

The method effectively handles label ambiguity and weak supervision.

Extensive experiments validate the robustness of HopS.

Abstract

Prompt learning has gained significant attention as a parameter-efficient approach for adapting large pre-trained vision-language models to downstream tasks. However, when only partial labels are available, its performance is often limited by label ambiguity and insufficient supervisory information. To address this issue, we propose Holistic Optimal Label Selection (HopS), leveraging the generalization ability of pre-trained feature encoders through two complementary strategies. First, we design a local density-based filter that selects the top frequent labels from the nearest neighbors' candidate sets and uses the softmax scores to identify the most plausible label, capturing structural regularities in the feature space. Second, we introduce a global selection objective based on optimal transport that maps the uniform sampling distribution to the candidate label distributions across a batch. By minimizing the expected transport cost, it can determine the most likely label assignments. These two strategies work together to provide robust label selection from both local and global perspectives. Extensive experiments on eight benchmark datasets show that HopS consistently improves performance under partial supervision and outperforms all baselines. Those results highlight the merit of holistic label selection and offer a practical solution for prompt learning in weakly supervised settings.