Search papers, labs, and topics across Lattice.
This paper introduces Domain Compensation (DOCO), a novel framework for Open-set Continual Test-Time Adaptation (OCTTA) that addresses both distributional and semantic shifts during inference. DOCO dynamically splits samples into ID and OOD, then learns a domain compensation prompt by aligning feature statistics of ID samples with the source domain while preserving semantic structure. Experiments on challenging benchmarks demonstrate that DOCO achieves state-of-the-art performance in OCTTA by synergistically performing domain adaptation and OOD detection.
Domain shifts and novel classes at test time can be tamed by nudging features back towards the source distribution, even for out-of-distribution examples.
Test-Time Adaptation (TTA) aims to mitigate distributional shifts between training and test domains during inference time. However, existing TTA methods fall short in the realistic scenario where models face both continually changing domains and the simultaneous emergence of unknown semantic classes, a challenging setting we term Open-set Continual Test-Time Adaptation (OCTTA). The coupling of domain and semantic shifts often collapses the feature space, severely degrading both classification and out-of-distribution detection. To tackle this, we propose DOmain COmpensation (DOCO), a lightweight and effective framework that robustly performs domain adaptation and OOD detection in a synergistic, closed loop. DOCO first performs dynamic, adaptation-conditioned sample splitting to separate likely ID from OOD samples. Then, using only the ID samples, it learns a domain compensation prompt by aligning feature statistics with the source domain, guided by a structural preservation regularizer that prevents semantic distortion. This learned prompt is then propagated to the OOD samples within the same batch, effectively isolating their semantic novelty for more reliable detection. Extensive experiments on multiple challenging benchmarks demonstrate that DOCO outperforms prior CTTA and OSTTA methods, establishing a new state-of-the-art for the demanding OCTTA setting.
