analiza aiqbench
This study systematically evaluates test-time adaptation (TTA) methods for EEG foundation models, revealing that directly applying existing techniques often degrades performance due to the unique challenges of neural signal distribution shifts. The findings highlight that optimization-free approaches offer greater stability than gradient-based methods in this domain, underscoring the necessity for specialized adaptation strategies rather than generic transfers from other fields.
Test-Time Adaptation for EEG Foundation Models: A Systematic Study under Real-World Distribution Shifts
arXiv:2604.16926v3 Announce Type: replace-cross Abstract: Electroencephalography (EEG) foundation models have shown strong potential for learning generalizable representations from large-scale neural data, yet their clinical deployment is hindered by distribution shifts across clinical settings, devices, and populations. Test-time adaptation (TTA) offers a promising solution by enabling models to adapt to unlabeled target data during inference without access to source data, a valuable property in healthcare settings constrained by privacy regulations and limited labeled data. However, its effectiveness for EEG remains largely underexplored. In this work, we introduce NeuroAdapt-Bench, a systematic benchmark for evaluating test-time adaptation methods on EEG foundation models under realistic distribution shifts. We evaluate representative TTA approaches from other domains across multiple pretrained foundation models, diverse downstream tasks, and heterogeneous datasets spanning in-distribution, out-of-distribution, and extreme modality shifts (e.g., Ear-EEG). Our results show that the evaluated TTA methods yield inconsistent gains and often degrade performance, with gradient-based approaches particularly prone to heavy degradation and optimization-free methods showing greater stability. For the evaluated EEG foundation models and representative TTA methods, these findings highlight the limitations of directly applying existing TTA techniques to EEG and underscore the need for domain-specific adaptation strategies.
Abstract:Electroencephalography (EEG) foundation models have shown strong potential for learning generalizable representations from large-scale neural data, yet their clinical deployment is hindered by distribution shifts across clinical settings, devices, and populations. Test-time adaptation (TTA) offers a promising solution by enabling models to adapt to unlabeled target data during inference without access to source data, a valuable property in healthcare settings constrained by privacy regulations and limited labeled data. However, its effectiveness for EEG remains largely underexplored. In this work, we introduce NeuroAdapt-Bench, a systematic benchmark for evaluating test-time adaptation methods on EEG foundation models under realistic distribution shifts. We evaluate representative TTA approaches from other domains across multiple pretrained foundation models, diverse downstream tasks, and heterogeneous datasets spanning in-distribution, out-of-distribution, and extreme modality shifts (e.g., Ear-EEG). Our results show that the evaluated TTA methods yield inconsistent gains and often degrade performance, with gradient-based approaches particularly prone to heavy degradation and optimization-free methods showing greater stability. For the evaluated EEG foundation models and representative TTA methods, these findings highlight the limitations of directly applying existing TTA techniques to EEG and underscore the need for domain-specific adaptation strategies.
Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Signal Processing (eess.SP)
arXiv:2604.16926 [cs.LG]
(or arXiv:2604.16926v3 [cs.LG] for this version)
https://doi.org/10.48550/arXiv.2604.16926 Focus to learn more arXiv-issued DOI via DataCite
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Text integral sub licența deschisă a publicației · 284 cuvinte
Subiecte conexe
Presentation: Building Evals for AI Adoption: from Principles to Practice
InfoQ · 29 mai 2026
What Drives Test-Time Adaptation for CLIP? A Controlled Empirical Study from an Update Perspective
arXiv · 7 aug. 2026
What Drives Test-Time Adaptation for CLIP? A Controlled Empirical Study from an Update Perspective
arXiv · 15 iun. 2026
Build intelligent security for healthcare APIs with Amazon Bedrock
AWS · 20 aug. 2026