Controllable Continual Test-Time Adaptation
This addresses the challenge of adapting models to continuously changing test conditions without source data, which is crucial for real-world AI applications, though it appears incremental by building on existing CTTA methods.
The paper tackles the problem of error accumulation in Continual Test-Time Adaptation (CTTA) due to uncontrollable domain shifts, and proposes a method that guides shifts to prevent category encroachment, resulting in improved performance as demonstrated by quantitative experiments and qualitative analyses like t-SNE plots.
Continual Test-Time Adaptation (CTTA) is an emerging and challenging task where a model trained in a source domain must adapt to continuously changing conditions during testing, without access to the original source data. CTTA is prone to error accumulation due to uncontrollable domain shifts, leading to blurred decision boundaries between categories. Existing CTTA methods primarily focus on suppressing domain shifts, which proves inadequate during the unsupervised test phase. In contrast, we introduce a novel approach that guides rather than suppresses these shifts. Specifically, we propose $\textbf{C}$ontrollable $\textbf{Co}$ntinual $\textbf{T}$est-$\textbf{T}$ime $\textbf{A}$daptation (C-CoTTA), which explicitly prevents any single category from encroaching on others, thereby mitigating the mutual influence between categories caused by uncontrollable shifts. Moreover, our method reduces the sensitivity of model to domain transformations, thereby minimizing the magnitude of category shifts. Extensive quantitative experiments demonstrate the effectiveness of our method, while qualitative analyses, such as t-SNE plots, confirm the theoretical validity of our approach.