Sebastian Lapuschkin

Technological University Dublin — Ireland

Computer Science · Physical Sciences

26h-index6.8kcitations129works1.92yr avg
OpenAlex↗ORCID↗Google Scholar↗Lab page↗

Accepting Students?

No reports yet. Know if this professor is taking students?

Research Topics

Explainable Artificial Intelligence (XAI)(60), Adversarial Robustness in Machine Learning(29), Machine Learning and Data Classification(20), Anomaly Detection Techniques and Applications(18), Advanced Neural Network Applications(17)

Publications129 total

Software for dataset-wide XAI: From local explanations to global insights with Zennit, CoRelAy, and ViRelAy
PLoS ONE·2026· 4 citedOpen Access
From Weights to Activations: Is Steering the Next Frontier of Adaptation?
2026Open Access
Deep learning-based multi project InP wafer simulation towards unsupervised surface defect detection
Journal of the Franklin Institute·2025Open Access
Atlas-Alignment: Making Interpretability Transferable Across Language Models
arXiv (Cornell University)·2025Open Access
Leveraging Influence Functions for Resampling Data in Physics-Informed Neural Networks
Communications in computer and information science·2025Open Access
Post-hoc Concept Disentanglement: From Correlated to Isolated Concept Representations
Communications in computer and information science·2025Open Access
Strategic Dishonesty Can Undermine AI Safety Evaluations of Frontier LLMs
arXiv (Cornell University)·2025Open Access
Iterative Inference in a Chess-Playing Neural Network
arXiv (Cornell University)·2025Open Access
Dyslexify: A Mechanistic Defense Against Typographic Attacks in CLIP
ArXiv.org·2025Open Access
Ensuring medical AI safety: interpretability-driven detection and mitigation of spurious model behavior and associated data
Machine Learning·2025· 4 citedOpen Access
Exploring the clinical value of concept-based AI explanations in gastrointestinal disease detection
Scientific Reports·2025Open Access
Künstliche Intelligenz auf dem Prüfstand: Anforderungen, Qualitätskriterien und Prüfwerkzeuge für medizinische Anwendungen
Bundesgesundheitsblatt - Gesundheitsforschung - Gesundheitsschutz·2025Open Access
Attribution-guided Pruning for Compression, Circuit Discovery, and Targeted Correction in LLMs
arXiv (Cornell University)·2025Open Access
Deep Learning-based Multi Project InP Wafer Simulation for Unsupervised Surface Defect Detection
arXiv (Cornell University)·2025Open Access
Relevance-driven Input Dropout: an Explanation-guided Regularization Technique
arXiv (Cornell University)·2025Open Access
From What to How: Attributing CLIP's Latent Components Reveals Unexpected Semantic Reliance
arXiv (Cornell University)·2025Open Access
The Atlas of In-Context Learning: How Attention Heads Shape In-Context Retrieval Augmentation
arXiv (Cornell University)·2025Open Access
See What I Mean? CUE: A Cognitive Model of Understanding Explanations
arXiv (Cornell University)·2025Open Access
Prisma: An Open Source Toolkit for Mechanistic Interpretability in Vision and Video
arXiv (Cornell University)·2025Open Access
ASIDE: Architectural Separation of Instructions and Data in Language Models
arXiv (Cornell University)·2025Open Access
page 1 of 7Next →

Frequent Co-authors

Wojciech Samek(48), Klaus‐Robert Müller(23), Alexander Binder(19), Grégoire Montavon(14), Christopher J. Anders(5), Reduan Achtibat(5), Leander Weber(5), Fabian Horst(5), Maximilian Dreyer(4), Thomas Wiegand(4), Djordje Slijepčević(4), Brian Horsak(4), Anna-Maria Raberger(4), Christian Breiteneder(4), Wolfgang I. Schöllhorn(4), Matthias Zeppelzauer(4), Jiamei Sun(4), Johanna Vielhaben(4), Sebastian Bosse(3), Sören Becker(3)