Progressive Hybrid Pseudo-Labeling for Unsupervised Domain Adaptation With Ascending Low-Rank Adaptation
IEEE Transactions on Image Processing · 2026 · 35 · 7332-7346
DOI: 10.1109/tip.2026.3709466 ↗Kurz gefasst
Fachartikel in der Zeitschrift IEEE Transactions on Image Processing, 2026. Verfasst von Yangtao Wang und 4 weiteren Autor:innen. Die Arbeit nennt 65 Literaturangaben. Gefördert von: National Natural Science Foundation of China, Guangdong Basic and Applied Basic Research Foundation, National Key Research and Development Program of China.
Abstract (Originalsprache)
Unsupervised domain adaptation (UDA) based on large vision-language models (VLMs) has recently demonstrated strong generalization ability, yet it remains fundamentally challenged by noisy pseudo-labels and inefficient adaptation under large domain shifts. In this paper, we propose Progressive Hybrid Pseudo-Labeling for UDA with Ascending Low-Rank Adaptation (termed as PHPL), a parameter-efficient paradigm that addresses these challenges from two complementary perspectives. 1) We introduce a progressive hybrid pseudo-labeling strategy that constructs target-domain supervision by fusing predictions from a frozen teacher model and an adaptive student model with a progressive weighting scheme. By gradually transferring predictive responsibility from the teacher to the student during training, PHPL effectively mitigates early-stage pseudo-label noise and stabilizes self-training under large domain shifts. 2) To enable efficient and stable adaptation of large VLMs, we propose an ascending low-rank adaptation strategy that allocates LoRA capacity in a depth-aware manner. Specifically, larger low-rank updates are assigned to deeper, semantically richer layers, while shallow layers remain lightly parameterized, striking a favorable balance between parameter efficiency and representational expressiveness. We conduct extensive experiments on five widely-used UDA benchmarks, including Office-Home, Office-31, VisDA-2017, Mini-DomainNet, and DomainNet. Experimental results verify that PHPL consistently achieves higher performance across various cross-domain scenarios compared with existing CNN, Transformer, and VLMs-based solutions. Notably, PHPL demonstrates strong robustness on highly challenging large-scale conditions while requiring significantly less computational overhead, validating the effectiveness and scalability of the proposed lightweight adaptation paradigm. The code is available at https://github.com/el2k/PHPL.
Förderung
- National Natural Science Foundation of China
- Guangdong Basic and Applied Basic Research Foundation
- National Key Research and Development Program of China
- Guangzhou Science and Technology Planning Project
- Guangdong Provincial Department of Education Innovation Strong School Youth Innovation Talent Project
Zitieren
APA
Wang, Y., Huang, M., Deng, X., Xie, Y., & Li, X. (2026). Progressive Hybrid Pseudo-Labeling for Unsupervised Domain Adaptation With Ascending Low-Rank Adaptation. IEEE Transactions on Image Processing, 35, 7332-7346. https://doi.org/10.1109/tip.2026.3709466
Vancouver
Wang Y, Huang M, Deng X, Xie Y, Li X. Progressive Hybrid Pseudo-Labeling for Unsupervised Domain Adaptation With Ascending Low-Rank Adaptation. IEEE Transactions on Image Processing. 2026;35:7332-7346. doi:10.1109/tip.2026.3709466
BibTeX
@article{Wang2026,
author = {Wang, Yangtao and Huang, Mingxin and Deng, Xingwei and Xie, Yanzhao and Li, Xiaocui},
title = {Progressive Hybrid Pseudo-Labeling for Unsupervised Domain Adaptation With Ascending Low-Rank Adaptation},
journal = {IEEE Transactions on Image Processing},
year = {2026},
volume = {35},
pages = {7332--7346},
doi = {10.1109/tip.2026.3709466}
}
EUVIMED