Metraipolis › archive › dc_925e8f29975b

On the Interaction Between Model Compression and Test-Time Adaptation — arXiv:2609.03604

Source
arXiv (en)
Original
https://arxiv.org/abs/2609.03604
License (as archived)
CC0-1.0
Archived
2026-10-17T01:11:58.916Z
SHA-256 of the archived text
b773907869f0eeb44e52b47cb92126aa4c2e472037f47891d176d993c25f1520
Stable URL of this page
https://metraipolis.com/documento/dc_925e8f29975b

How to cite

On the Interaction Between Model Compression and Test-Time Adaptation — arXiv:2609.03604 — arXiv (en), https://arxiv.org/abs/2609.03604 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_925e8f29975b (Metraipolis archive dc_925e8f29975b, sha256 b773907869f0eeb44e52b47cb92126aa4c2e472037f47891d176d993c25f1520)
BibTeX
@misc{dc925e8f29975b,
  title = {On the Interaction Between Model Compression and Test-Time Adaptation — arXiv:2609.03604},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.03604},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_925e8f29975b; sha256 b773907869f0eeb44e52b47cb92126aa4c2e472037f47891d176d993c25f1520},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.03604. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

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.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-04 in cs.AI).
Categorie arXiv: Machine Learning (cs.LG) ; Artificial Intelligence (cs.AI).
Pagina del lavoro: https://arxiv.org/abs/2609.03604 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito).
Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina.
Letto il 2026-10-17T00:28:02.983Z dalla pagina https://arxiv.org/catchup/cs.AI/2026-09-04?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).