LanDA: Language-Guided Multi-Source Domain Adaptation

Multi-Source Domain Adaptation (MSDA) aims to mitigate changes in data distribution when transferring knowledge from multiple labeled source domains to an unlabeled target ___domain. However, existing MSDA techniques assume target ___domain images are available, yet overlook image-rich semantic information. Consequently, an open question is whether MSDA can be guided solely by textual cues in the absence of target ___domain images. By employing a multimodal model with a joint image and language embedding space, we propose a novel language-guided MSDA approach, termed LanDA, based on optimal transfer theory, which facilitates the transfer of multiple source domains to a new target ___domain, requiring only a textual description of the target ___domain without needing even a single target ___domain image, while retaining task-relevant information. We present extensive experiments across different transfer scenarios using a suite of relevant benchmarks, demonstrating that LanDA outperforms standard fine-tuning and ensemble approaches in both target and source domains.