Banner Banner

SEARCH

ALL PUBLICATIONS

Maximilian von Klinski, Sebastian Lapuschkin, Wojciech Samek, Lennart Bürger

Stress-Testing LLM Lie Detectors: Role-Play Failures and Spurious Correlations

September 30, 2026
https://doi.org/10.48550/arXiv.2609.39807

Ankur Singha, Jacob Finkenrath, Karl Jansen, Vipul Arora, Shinichi Nakajima

Multilevel Plaquette-Space Sampling for Lattice Gauge Theories with Local Constraint Solves

September 28, 2026
https://doi.org/10.48550/arXiv.2609.35722

Jakob Steglich, Justus Meyer zu Bexten, Shakiba Moradi, Laure Ciernik, Simon M. Hofmann, Mina Jamshidi Idaji

Representational and Functional Robustness to Electrode Montages in EEG Foundation Models

September 28, 2026
https://doi.org/10.48550/arXiv.2609.36288

Ping Xiong, Shanglin Li, Yi Ding, Thomas Schnake, Shinichi Nakajima

Explaining Hyperbolic Neural Networks via Geometry-Aware Relevance Propagation

September 28, 2026
https://doi.org/10.48550/arXiv.2609.35128

Sanaz Karimijafarbigloo, Amirhossein Kazerouni, Alireza Kheirkhah, Alireza Kheirkhah, Sergey Protserov, Reza Azad, Michael Brudno, Babak Taati, Wojciech Samek, Dorit Merhof et al.

Probabilistic Multi-Rater Segmentation via Style-Aware Boundary Conditioning

September 27, 2026
https://papers.miccai.org/miccai-2026/paper/5829_paper.pdf

Denis Shcherba, Adrian Abel, Eckart Cobo-Briesewitz, Paul Mattes, Wojciech Samek, Marc Toussaint

Estimate, Don't Imitate: Reusing Differentiable State-Based Policies for Visuomotor Control

September 27, 2026
https://doi.org/10.48550/arXiv.2609.34018

Moritz Weckbecker, Sweta Jena, Jonas Müller, Ponnurangam Kumaraguru, Sebastian Lapuschkin, Wojciech Samek, Louis Jaburi, Gonçalo Paulo

Can Data Attribution Filter Out Subliminal Learning? Not Reliably

September 17, 2026
https://doi.org/10.48550/arXiv.2609.20027

Marcel F. Langer, Adrian Hill, Michele Ceriotti

Truncated automatic sparse differentiation for machine learning interatomic potentials

September 17, 2026
Machine learning interatomic potentials (MLIPs) learn the mapping from atomic positions to potential energy. The forces, the negative gradient of this energy, drive molecular dynamics and are readily obtained using automatic differentiation. Higher-order