STAMP: Spatial-Temporal Adapter with Multi-Head Pooling

Brad Shook, Abby Turner, Jieshi Chen, Michał Wiliński, Mononito Goswami +2 more
11/13/2025
cs.LGcs.AI

No official code found. Best implementation: autonlab/STAMP (6 stars, MIT, last commit 2026-01-23).

Abstract

Time series foundation models (TSFMs) pretrained on data from multiple domains have shown strong performance on diverse modeling tasks. Various efforts have been made to develop foundation models specific to electroencephalography (EEG) data, which records brain electrical activity as time series. However, no comparative analysis of EEG-specific foundation models (EEGFMs) versus general TSFMs has been performed on EEG-specific tasks. We introduce a novel Spatial-Temporal Adapter with Multi-Head Pooling (STAMP), which leverages univariate embeddings produced by a general TSFM, implicitly models spatial-temporal characteristics of EEG data, and achieves performance comparable to state-of-the-art EEGFMs. A comprehensive analysis is performed on 8 benchmark datasets of clinical tasks using EEG for classification, along with ablation studies. Our proposed adapter is lightweight in trainable parameters and flexible in the inputs it can accommodate, supporting easy modeling of EEG data using TSFMs.

View on arXivView PDF

Code Implementations(1)

autonlab/STAMPCommunity66%
MIT

Official Implementation of "STAMP: Spatial-Temporal Adapter with Multi-Head Pooling". arXiv: https://www.arxiv.org/abs/2511.10848

60Shell, PythonSep 8, 20259 months agoMIT

Similar Papers

Are foundation models useful feature extractors for electroencephalography analysis?

86% match2025

Leveraging Generic Time Series Foundation Models for EEG Classification

85% match2025

Are EEG Sequences Time Series? EEG Classification with Time Series Models and Joint Subject Training

84% match2024

FoME: A Foundation Model for EEG using Adaptive Temporal-Lateral Attention Scaling

82% match2024

ALFEE: Adaptive Large Foundation Model for EEG Representation

82% match1 repo32025

NeurIPT: Foundation Model for Neural Interfaces

81% match1 repo112025

Cite this paper

@article{shook2025stamp,
  title  = {STAMP: Spatial-Temporal Adapter with Multi-Head Pooling},
  author = {Brad Shook and Abby Turner and Jieshi Chen and Michał Wiliński and Mononito Goswami and Jonathan Elmer and Artur Dubrawski},
  year   = {2025},
  eprint = {2511.10848},
  archivePrefix = {arXiv},
  url    = {http://arxiv.org/abs/2511.10848v2}
}

Add a badge to your README

Shows whether official code is linked for this paper.

Code of Paper badge
[![Code of Paper](https://codeofpaper.com/badge/2511.10848.svg)](https://codeofpaper.com/paper/2511.10848)

Discussion