MITA: Bridging the Gap between Model and Data for Test-time Adaptation
This addresses the challenge of enhancing model generalizability in complex real-world scenarios like outliers or mixed distributions, representing a novel paradigm shift rather than an incremental improvement.
The paper tackles the problem of suboptimal performance in Test-Time Adaptation (TTA) methods due to over-reliance on batch-level statistics, proposing MITA to bridge the gap between model and data distributions, and demonstrates superior performance over state-of-the-art methods in experiments across outlier, mixture, and pure scenarios.
Test-Time Adaptation (TTA) has emerged as a promising paradigm for enhancing the generalizability of models. However, existing mainstream TTA methods, predominantly operating at batch level, often exhibit suboptimal performance in complex real-world scenarios, particularly when confronting outliers or mixed distributions. This phenomenon stems from a pronounced over-reliance on statistical patterns over the distinct characteristics of individual instances, resulting in a divergence between the distribution captured by the model and data characteristics. To address this challenge, we propose Meet-In-The-Middle based Test-Time Adaptation ($\textbf{MITA}$), which introduces energy-based optimization to encourage mutual adaptation of the model and data from opposing directions, thereby meeting in the middle. MITA pioneers a significant departure from traditional approaches that focus solely on aligning the model to the data, facilitating a more effective bridging of the gap between model's distribution and data characteristics. Comprehensive experiments with MITA across three distinct scenarios (Outlier, Mixture, and Pure) demonstrate its superior performance over SOTA methods, highlighting its potential to significantly enhance generalizability in practical applications.