Decentralized Attention Fails Centralized Signals: Rethinking Transformers for Medical Time Series

ICLROral2026

Authors: Guoqi Yu, Juncheng Wang, Chen Yang, Jing Qin, Angelica I Aviles-Rivero, Shujun Wang
Venue: ICLR 2026
Track: Oral

TL;DR

We propose a centralized module to replace decentralized attention in Transformer for centralized medical time series like EEG and ECG.

Opening excerpt from the authors’ abstract. source

Read the paper

OpenReview

Topics

time series transformer attention

← All ICLR 2026 Oral papers · Browse the whole archive