SNAP: Low-Latency Test-Time Adaptation with Sparse Updates

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Cha, Hyeongheon, Kim, Dong Min, Chung, Hye Won, Gong, Taesik, Lee, Sung-Ju
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866917092364648448
author Cha, Hyeongheon
Kim, Dong Min
Chung, Hye Won
Gong, Taesik
Lee, Sung-Ju
author_facet Cha, Hyeongheon
Kim, Dong Min
Chung, Hye Won
Gong, Taesik
Lee, Sung-Ju
contents Test-Time Adaptation (TTA) adjusts models using unlabeled test data to handle dynamic distribution shifts. However, existing methods rely on frequent adaptation and high computational cost, making them unsuitable for resource-constrained edge environments. To address this, we propose SNAP, a sparse TTA framework that reduces adaptation frequency and data usage while preserving accuracy. SNAP maintains competitive accuracy even when adapting based on only 1% of the incoming data stream, demonstrating its robustness under infrequent updates. Our method introduces two key components: (i) Class and Domain Representative Memory (CnDRM), which identifies and stores a small set of samples that are representative of both class and domain characteristics to support efficient adaptation with limited data; and (ii) Inference-only Batch-aware Memory Normalization (IoBMN), which dynamically adjusts normalization statistics at inference time by leveraging these representative samples, enabling efficient alignment to shifting target domains. Integrated with five state-of-the-art TTA algorithms, SNAP reduces latency by up to 93.12%, while keeping the accuracy drop below 3.3%, even across adaptation rates ranging from 1% to 50%. This demonstrates its strong potential for practical use on edge devices serving latency-sensitive applications. The source code is available at https://github.com/chahh9808/SNAP.
format Preprint
id arxiv_https___arxiv_org_abs_2511_15276
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle SNAP: Low-Latency Test-Time Adaptation with Sparse Updates
Cha, Hyeongheon
Kim, Dong Min
Chung, Hye Won
Gong, Taesik
Lee, Sung-Ju
Machine Learning
Test-Time Adaptation (TTA) adjusts models using unlabeled test data to handle dynamic distribution shifts. However, existing methods rely on frequent adaptation and high computational cost, making them unsuitable for resource-constrained edge environments. To address this, we propose SNAP, a sparse TTA framework that reduces adaptation frequency and data usage while preserving accuracy. SNAP maintains competitive accuracy even when adapting based on only 1% of the incoming data stream, demonstrating its robustness under infrequent updates. Our method introduces two key components: (i) Class and Domain Representative Memory (CnDRM), which identifies and stores a small set of samples that are representative of both class and domain characteristics to support efficient adaptation with limited data; and (ii) Inference-only Batch-aware Memory Normalization (IoBMN), which dynamically adjusts normalization statistics at inference time by leveraging these representative samples, enabling efficient alignment to shifting target domains. Integrated with five state-of-the-art TTA algorithms, SNAP reduces latency by up to 93.12%, while keeping the accuracy drop below 3.3%, even across adaptation rates ranging from 1% to 50%. This demonstrates its strong potential for practical use on edge devices serving latency-sensitive applications. The source code is available at https://github.com/chahh9808/SNAP.
title SNAP: Low-Latency Test-Time Adaptation with Sparse Updates
topic Machine Learning
url https://arxiv.org/abs/2511.15276