C-DGPA: Class-Centric Dual-Alignment Generative Prompt Adaptation
This work addresses domain adaptation for vision-language models, improving performance in tasks like image classification across different domains, but it is incremental as it builds on existing prompt-tuning strategies.
The paper tackles the challenge of domain discrepancies in unsupervised domain adaptation for vision-language models by proposing C-DGPA, which aligns both marginal and conditional distributions through a dual-branch architecture, achieving state-of-the-art results on benchmarks like OfficeHome, Office31, and VisDA-2017.
Unsupervised Domain Adaptation transfers knowledge from a labeled source domain to an unlabeled target domain. Directly deploying Vision-Language Models (VLMs) with prompt tuning in downstream UDA tasks faces the signifi cant challenge of mitigating domain discrepancies. Existing prompt-tuning strategies primarily align marginal distribu tion, but neglect conditional distribution discrepancies, lead ing to critical issues such as class prototype misalignment and degraded semantic discriminability. To address these lim itations, the work proposes C-DGPA: Class-Centric Dual Alignment Generative Prompt Adaptation. C-DGPA syner gistically optimizes marginal distribution alignment and con ditional distribution alignment through a novel dual-branch architecture. The marginal distribution alignment branch em ploys a dynamic adversarial training framework to bridge marginal distribution discrepancies. Simultaneously, the con ditional distribution alignment branch introduces a Class Mapping Mechanism (CMM) to align conditional distribu tion discrepancies by standardizing semantic prompt under standing and preventing source domain over-reliance. This dual alignment strategy effectively integrates domain knowl edge into prompt learning via synergistic optimization, ensur ing domain-invariant and semantically discriminative repre sentations. Extensive experiments on OfficeHome, Office31, and VisDA-2017 validate the superiority of C-DGPA. It achieves new state-of-the-art results on all benchmarks.