collaborators

5 papers

cs.CL2026

Steering LLMs? Actually, Sparse Autoencoders can outperform simple baselines

Mikkel Godsk Jørgensen, Lars Kai Hansen

Sparse Autoencoders (SAEs) have been seen as a promising avenue for exploring the internals of Large Language Models (LLMs) and for steering model output generation. When AxBench -…

cs.LG2026

Mechanistic Interpretability of EEG Foundation Models via Sparse Autoencoders

William Lehn-Schiøler, Magnus Ruud Kjær, Rahul Thapa +10

EEG foundation models achieve state-of-the-art clinical performance, yet the internal computations driving their predictions remain opaque: a barrier to clinical trust. We apply To…

cs.LG2026

Pretraining on Sleep Data Improves non-Sleep Biosignal Tasks

William Lehn-Schiøler, Magnus Ruud Kjær, Phillip Hempel +7

Sleep foundation models have recently demonstrated strong performance on in-domain polysomnography tasks, including sleep staging, apnea detection, and disease risk prediction. In…

cs.CV2025

From Colors to Classes: Emergence of Concepts in Vision Transformers

Teresa Dorszewski, Lenka Tětková, Robert Jenssen +2

Vision Transformers (ViTs) are increasingly utilized in various computer vision tasks due to their powerful representation capabilities. However, it remains understudied how ViTs p…

eess.AS2025

How Redundant Is the Transformer Stack in Speech Representation Models?

Teresa Dorszewski, Albert Kjøller Jacobsen, Lenka Tětková +1

Self-supervised speech representation models, particularly those leveraging transformer architectures, have demonstrated remarkable performance across various tasks such as speech…