Unsupervised Domain Adaptation (UDA) has emerged as a pivotal technique for enhancing machine learning models' performance in unlabeled target domains with domain shifts. This technique is fundamentally achieved by aligning the domain distributions of source and target domains within a latent feature space, thereby enhancing model robustness across heterogeneous data distributions. However, the inherent discrepancy between source and target domain distributions poses significant challenges in identifying the optimal latent space. Furthermore, projecting both domains into suboptimal latent spaces may induce substantial semantic information loss, particularly compromising discriminative feature representations critical for final tasks. In this paper, Our systematic analysis reveals that natural language representations inherently possess stronger semantic abstraction capabilities than visual features in natural images. As a result, natural language tends to have a smaller domain shifts. Motivated by this discovery, we propose a novel model that systematically transforms visual patterns into structured linguistic representations. This cross-modal translation mechanism leverages the invariant semantic properties of natural language to mitigate domain shifts while preserving task-critical semantic hierarchies. Our model leverages the inherent abstraction capacity of linguistic structures to enhance cross-domain generalization, effectively bridging the visual-semantic gap in unsupervised adaptation scenarios. Our model comprises three core components: (1) a text classification branch translating images to text for prediction, (2) an image adaptation branch supplementing visual details, and (3) an ensemble mechanism reconciling text abstraction with visual granularity through mismatch detection. Extensive experiments on three benchmark datasets validate the effectiveness of our model, achieving state-of-the-art performance.
Jia et al. (Thu,) studied this question.