arXiv AI September 4, 2026 By Francesco Corti, Dong Wang, Young D. Kwon, Cecilia Mascolo, Olga Saukh

On the Interaction Between Model Compression and Test-Time Adaptation

The paper investigates how model compression impacts test-time adaptation (TTA) in deep neural networks. Using ResNet‑18 and ViT‑Base on CIFAR‑10‑C and ImageNet‑C, the authors evaluate several compression techniques alongside standard TTA methods, introducing a diagnostic framework to assess representational expressivity and adaptation subspace compatibility. Results show that while compressed models maintain high accuracy under supervised adaptation, their TTA performance deteriorates with increased compression due to reduced representational diversity and structural constraints that limit recoverability.

Machine-generated by The Flow from the publisher's headline and feed description — not written or checked by a human. The full article lives at arXiv AI.

efficiency

arXiv Machine Learning

Sep 22

SPeaR: Test-Time Adaptation with Steering Primitives for Realigning Representations

arXiv:2609.24111v1 Announce Type: new Abstract: Test-time adaptation (TTA) addresses distribution shift using only unlabeled test data. Existing methods typically adapt pretrained models by updating...

By Muhammad Sudipto Siam Dip, Ali Etemad

llms

arXiv AI

Aug 18

Fast Test-Time Refinement for Robust Learned Image Compression

arXiv:2608. 15113v1 Announce Type: cross Abstract: Learned image compression (LIC) has demonstrated remarkable rate-distortion (RD) performance in benign settings.

By Jiaming Liang, Chi-Man Pun, Weisi Lin

safety

arXiv AI

Jun 9

DOME: Learning Transferable Domain Variables from Sparse Supervision for Test-Time Adaptation

arXiv:2606. 07646v1 Announce Type: cross Abstract: Test-time adaptation (TTA) aims to align a model to shifting test domains using only unlabeled streaming data.

By Xiaoran Xu, Yifan Xu, Yupeng Wu, Xiaoshan Yang, Changsheng Xu

multimodal benchmarks

arXiv Machine Learning

Aug 5

Test-Time Augmentation for Tabular-to-Image Classifiers under Distribution Shifts

arXiv:2608. 03557v1 Announce Type: cross Abstract: Tabular-to-image methods that convert tabular data into visual representations have emerged as a novel paradigm for leveraging the high performance of deep learning models.

By Malena Loza, Felipe Grijalva, Eva Milara, Luis Bote-Curiel, Francisco J. Lara-Abelenda, David Chushig-Muzo

computer-vision benchmarks

arXiv Computer Vision

Aug 24

Continual Test-Time Adaptation in Computer Vision: Methods, Benchmarks, and Future Directions

The paper surveys Continual Test-Time Adaptation (CTTA), a framework that adapts pretrained computer‑vision models to non‑stationary target distributions without source data or labeled targets, while avoiding catastrophic forgetting and error accumulation. It formally defines the CTTA problem, categorizes existing methods into optimization‑based, parameter‑efficient, and architecture‑based families, and reviews representative techniques and benchmarks across standard evaluation settings. The survey also outlines current limitations and proposes future research directions, such as adapting foundation models and black‑box systems.

By Sarthak Kumar Maharana, Shambhavi Mishra, Yunbei Zhang, Shuaicheng Niu, Taki Hasan Rafi, Jihun Hamm, Marco Pedersoli, Jose Dolz, Yunhui Guo

llms computer-vision benchmarks

arXiv Computer Vision

Aug 27

LongVU-TTT: Causal Test-Time Training for Visual Resampling in Long Video Understanding

LongVU‑TTT is a causal test‑time training method for long‑video multimodal large language models that inserts a convolutional resampler with fast‑weight updates between the vision encoder and the LLM. The fast weights adapt per video and contextualize frame features before compression, while a hybrid selector keeps explicit visual evidence for downstream reasoning. Experiments show that TTT‑Conv outperforms TTT‑MLP and bidirectional Mamba2 on MLVU, and beats attention‑ and fixed‑state recurrent resamplers on three benchmarks, achieving competitive results on five video‑understanding tasks after reducing 512 frames to 128 LLM frames.

By Mahmoud Ahmed, Sameh Abdulah, Olatunji Ruwase, Sam Ade Jacobs, Mathis Bode, Mohamed Elhoseiny

llms benchmarks