AI Paper Atlas

Research map / Time series / Foundation models for time series

Brain-Semantoks: Learning Semantic Tokens of Brain Dynamics with a Self-Distilled Foundation Model

Sam Gijsen, Marc-Andre Schulz, Kerstin Ritter

ICLR 2026 · Time series › Foundation models for time series · approach: Time series classification · 2 citations (OpenAlex)

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
self-distillation with semantic tokenizer instead of reconstruction-based pretraining (linear probe)BrainLM / Brain-JEPA · ABIDEbalanced accuracy: 53.84 -> 65.13Improved
sameBrain-JEPA · SRPBS SZbalanced accuracy: 57.63 -> 69.26Improved
sameBrainLM · HBN Agebalanced accuracy: 43.89 -> 39.16Hurt
add the global CLS self-distillation loss (ablation)JEPA-style baseline · 10-task averagebalanced accuracy: 47.32 -> 52.39Improved
add Teacher-guided Temporal Regularizer for first 5% of training (ablation)Brain-Semantoks · 10-task averagebalanced accuracy: 50.88 -> 52.39Improved
keep TTR for the entire pretraining instead of 5% (ablation)Brain-Semantoks · 10-task averagebalanced accuracy: 52.39 -> 49.60Hurt
use slice masking instead of random masking (ablation)Brain-Semantoks · 10-task averagebalanced accuracy: 51.03 -> 52.39Improved
use block masking instead of slice masking (ablation)Brain-Semantoks · 10-task averagebalanced accuracy: 52.39 -> 49.71Hurt
use structured dual-branch tokenizer kernel instead of single dense kernel 20 (ablation)Brain-Semantoks · 10-task averagebalanced accuracy: 48.28 -> 52.39Improved
set token loss weight lambda_Tok to 0.5 instead of 1.0 (ablation)Brain-Semantoks · 10-task averagebalanced accuracy: 51.62 -> 52.39Improved
semantic tokenizer instead of per-ROI linear projection (ablation)Brain-Semantoks · scaling (random init)linear probe accuracy: up to 12% gainImproved
more pretraining dataBrain-Semantoks · in- and out-of-distribution taskslinear probe accuracyImproved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

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