Problem
Research questions and friction points this paper is trying to address.
model compression
test-time adaptation
distribution shift
representational expressivity
adaptation subspace compatibility
On the Interaction Between Model Compression and Test-Time Adaptation
๐ค AI Summary
็ ็ฉถๅๆไบๆจกๅๅ็ผฉไธๆต่ฏๆถ้ๅบไน้ด็็ธไบไฝ็จ๏ผ้่ฟๅค็งๆนๆณ่ฏไผฐๅ็ฐๅ็ผฉไผ้ไฝๆจกๅ็้ๅบๆง๏ผ้่ฎพ่ฎกไฟๆ้ๅบๆง็ๅ็ผฉ็ญ็ฅใ
๐ Abstract
Deep neural networks deployed in the wild must be both efficient and adaptable, requiring model compression and test-time adaptation (TTA). While both are well studied in isolation, their interaction remains poorly understood. We systematically analyze how structured compression affects a model's ability to adapt under distribution shift. Using ResNet-18 and ViT-Base on CIFAR-10-C and ImageNet-C, we evaluate multiple compression methods combined with standard TTA techniques. We introduce a diagnostic framework that examines representational expressivity and adaptation subspace compatibility. Our results reveal a consistent gap: although compressed models retain high accuracy under supervised adaptation, their TTA performance degrades significantly with increasing compression. We show that this stems from reduced representational diversity and structural constraints that limit recoverability. These effects strongly depend on the compression method, highlighting the need to design compression strategies that preserve adaptability.
Innovation
Methods, ideas, or system contributions that make the work stand out.
model compression
test-time adaptation
representational expressivity
adaptation subspace compatibility
๐ Similar Papers
No similar papers found.
