analiza aiqbench
This paper introduces EEG-FM-Bench, a standardized evaluation framework designed to address critical gaps in comparing electroencephalography foundation models. By integrating diverse datasets and diagnostic tools for gradient analysis, it reveals that factors like objective alignment are more predictive of transfer performance than raw model scale alone. The release provides the necessary infrastructure for reproducible research and fair benchmarking within brain-computer interface AI systems.
EEG-FM-Bench: A Comprehensive Benchmark for the Systematic Evaluation and Diagnostic Analyses of EEG Foundation Models
arXiv:2508.17742v3 Announce Type: replace-cross Abstract: Electroencephalography foundation models (EEG-FMs) have advanced brain signal analysis, but the lack of standardized evaluation benchmarks impedes model comparison and scientific progress. Current evaluations rely on inconsistent protocols that render cross-model comparisons unreliable, while a lack of diagnostic analyses obscures the internal mechanisms driving transfer efficiency and scaling behaviors. To address this, we introduce \textbf{EEG-FM-Bench}, a unified system for the standardized evaluation of EEG-FMs. The benchmark integrates 14 datasets across 10 paradigms and incorporates diverse experimental settings, including multiple fine-tuning strategies, task organizations, and classifier configurations, supported by tools for gradient and representation analysis. Our experiments and analysis reveal several critical insights: (1) multi-task learning often acts as a useful regularizer that mitigates overfitting in data-scarce EEG contexts, although negative transfer can arise under specific task paradigms; (2) pre-training efficiency is currently limited by gradient conflicts between reconstruction objectives and downstream tasks; (3) under released checkpoints and a matched downstream protocol, model or data scale alone does not fully explain transfer performance, while objective alignment, adaptation compatibility, and EEG-specific design appear to be important factors. This benchmark enables fair comparison and reproducible analysis, providing a step toward fairer comparison and more interpretable analysis of EEG-FMs. Code is available at https://github.com/xw1216/EEG-FM-Bench.
Abstract:Electroencephalography foundation models (EEG-FMs) have advanced brain signal analysis, but the lack of standardized evaluation benchmarks impedes model comparison and scientific progress. Current evaluations rely on inconsistent protocols that render cross-model comparisons unreliable, while a lack of diagnostic analyses obscures the internal mechanisms driving transfer efficiency and scaling behaviors. To address this, we introduce \textbf{EEG-FM-Bench}, a unified system for the standardized evaluation of EEG-FMs. The benchmark integrates 14 datasets across 10 paradigms and incorporates diverse experimental settings, including multiple fine-tuning strategies, task organizations, and classifier configurations, supported by tools for gradient and representation analysis. Our experiments and analysis reveal several critical insights: (1) multi-task learning often acts as a useful regularizer that mitigates overfitting in data-scarce EEG contexts, although negative transfer can arise under specific task paradigms; (2) pre-training efficiency is currently limited by gradient conflicts between reconstruction objectives and downstream tasks; (3) under released checkpoints and a matched downstream protocol, model or data scale alone does not fully explain transfer performance, while objective alignment, adaptation compatibility, and EEG-specific design appear to be important factors. This benchmark enables fair comparison and reproducible analysis, providing a step toward fairer comparison and more interpretable analysis of EEG-FMs. Code is available at this https URL.
Signal Processing (eess.SP); Artificial Intelligence (cs.AI); Human-Computer Interaction (cs.HC)
arXiv:2508.17742 [eess.SP]
(or arXiv:2508.17742v3 [eess.SP] for this version)
Text integral sub licența deschisă a publicației · 219 cuvinte
Subiecte conexe
GlucoFM-Bench: Benchmarking Time-Series Foundation Models for Blood Glucose Forecasting
arXiv · 8 iun. 2026
WPBench: A Comprehensive Benchmark for Wind Power Forecasting
arXiv · 22 sept. 2026
Amazon Research Awards recipients announced
Amazon Science · 27 mai 2026
EMG-FM-Bench: A Comprehensive Benchmark for Foundation Model Transfer and Adaptation on Electromyography
arXiv · 6 oct. 2026