Adaptive Cache Enhancement for Test-Time Adaptation of Vision-Language Models

TL;DR

This paper introduces ACE, a novel cache-based test-time adaptation framework for vision-language models that dynamically constructs class-specific caches to improve robustness and accuracy under distribution shifts.

Contribution

ACE employs class-wise thresholds and iterative refinement to build a robust cache, enabling adaptive decision boundaries and enhanced out-of-distribution performance.

Findings

Achieves state-of-the-art results on 15 benchmarks.

Demonstrates superior robustness under distribution shifts.

Outperforms existing TTA methods in diverse scenarios.

Abstract

Vision-language models (VLMs) exhibit remarkable zero-shot generalization but suffer performance degradation under distribution shifts in downstream tasks, particularly in the absence of labeled data. Test-Time Adaptation (TTA) addresses this challenge by enabling online optimization of VLMs during inference, eliminating the need for annotated data. Cache-based TTA methods exploit historical knowledge by maintaining a dynamic memory cache of low-entropy or high-confidence samples, promoting efficient adaptation to out-of-distribution data. Nevertheless, these methods face two critical challenges: (1) unreliable confidence metrics under significant distribution shifts, resulting in error accumulation within the cache and degraded adaptation performance; and (2) rigid decision boundaries that fail to accommodate substantial distributional variations, leading to suboptimal predictions. To overcome these limitations, we introduce the Adaptive Cache Enhancement (ACE) framework, which constructs a robust cache by selectively storing high-confidence or low-entropy image embeddings per class, guided by dynamic, class-specific thresholds initialized from zero-shot statistics and iteratively refined using an exponential moving average and exploration-augmented updates. This approach enables adaptive, class-wise decision boundaries, ensuring robust and accurate predictions across diverse visual distributions. Extensive experiments on 15 diverse benchmark datasets demonstrate that ACE achieves state-of-the-art performance, delivering superior robustness and generalization compared to existing TTA methods in challenging out-of-distribution scenarios.