AI Paper Atlas

Research map / Label-efficient and robust learning / Recognition and long-tail

Quantifying the Generalization Gap in Seizure Detection: A Large-Scale Empirical Benchmark via the SzCORE Challenge

Jonathan Dan, Amirhossein Shahbazinia, Christodoulos Kechris, David Atienza

ICML 2026 · Label-efficient and robust learning › Recognition and long-tail · approach: Dense features and foundation priors

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
evaluate on the independent Filadelfia set instead of developers' self-selected datasets28 submitted seizure detectors · Filadelfiaevent-based F1Hurt
evaluate Sz Transformer on Filadelfia instead of SeizeIT1Sz Transformer · event-based F1event-based F1 (%): 36 -> 32Hurt
evaluate DeepSOZ-HEM on Filadelfia instead of SeizeIT1DeepSOZ-HEM · event-based F1event-based F1 (%): 24 -> 30Improved
evaluate SeizUnet on Filadelfia instead of SeizeIT1SeizUnet · event-based F1event-based F1 (%): 32 -> 11Hurt
evaluate eventNet on Filadelfia instead of SienaeventNet · event-based F1event-based F1 (%): 53 -> 14Hurt
evaluate zhu-transformer on Filadelfia instead of Sienazhu-transformer · event-based F1event-based F1 (%): 48 -> 19Hurt
raise classification threshold from S4Seizure v1 to v2 (ablation)S4Seizure · Filadelfiaevent-based F1 (%): 27 -> 27No clear effect
lower threshold from S4Seizure v2 to v3 (ablation)S4Seizure · Filadelfiaevent-based F1 (%): 27 -> 22Hurt
change HySEIZa architecture/threshold from v1 to v2 (ablation)HySEIZa · Filadelfiaevent-based F1 (%): 24 -> 14Hurt
use top mean-rank algorithm HySEIZa v1 instead of F1-leader Sz TransformerSz Transformer · Filadelfiaper-subject rank: Cliff's delta 0.01No clear effect
use top algorithms instead of classical Random ForestRandom Forest · Filadelfiaper-subject F1 rank: Cliff's delta > 0.6Improved
seizure shorter duration (median 48 s vs 118 s)seizure detectors · Filadelfiadetection rate: 48 s vs 118 sHurt

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy