Maximilian Dreyer

28 papers A* 3Journal 17Unranked 8
YearRankTypeTitle / Venue / Authors
2026 J jnl
X-SYS: A Reference Architecture for Interactive Explanation Systems.
CoRR
Tobias Labarta, Nhi Hoang, Maximilian Dreyer, Jim Berend, Oleg Hein, Jackie Ma, Wojciech Samek, Sebastian Lapuschkin
2025 J jnl
Attribution-guided Pruning for Compression, Circuit Discovery, and Targeted Correction in LLMs.
CoRR
Sayed Mohammad Vakilzadeh Hatefi, Maximilian Dreyer, Reduan Achtibat, Patrick Kahardipraja, Thomas Wiegand, Wojciech Samek, Sebastian Lapuschkin
2025 J jnl
From What to How: Attributing CLIP's Latent Components Reveals Unexpected Semantic Reliance.
CoRR
Maximilian Dreyer, Lorenz Hufe, Jim Berend, Thomas Wiegand, Sebastian Lapuschkin, Wojciech Samek
2025 J jnl
Mechanistic understanding and validation of large AI models with SemanticLens.
CoRR
Maximilian Dreyer, Jim Berend, Tobias Labarta, Johanna Vielhaben, Thomas Wiegand, Sebastian Lapuschkin, Wojciech Samek
2025 A* conf
Navigating Neural Space: Revisiting Concept Activation Vectors to Overcome Directional Divergence.
ICLR
Frederik Pahde, Maximilian Dreyer, Moritz Weckbecker, Leander Weber, Christopher J. Anders, Thomas Wiegand, Wojciech Samek, Sebastian Lapuschkin
2025 J jnl
Towards Mechanistic Defenses Against Typographic Attacks in CLIP.
CoRR
Lorenz Hufe, Constantin Venhoff, Maximilian Dreyer, Sebastian Lapuschkin, Wojciech Samek
2024 A* conf
AttnLRP: Attention-Aware Layer-Wise Relevance Propagation for Transformers.
ICML
Reduan Achtibat, Sayed Mohammad Vakilzadeh Hatefi, Maximilian Dreyer, Aakriti Jain, Thomas Wiegand, Sebastian Lapuschkin, Wojciech Samek
2024 J jnl
AttnLRP: Attention-Aware Layer-wise Relevance Propagation for Transformers.
CoRR
Reduan Achtibat, Sayed Mohammad Vakilzadeh Hatefi, Maximilian Dreyer, Aakriti Jain, Thomas Wiegand, Sebastian Lapuschkin, Wojciech Samek
2024 — conf
Explainable Concept Mappings of MRI: Revealing the Mechanisms Underlying Deep Learning-Based Brain Disease Classification.
xAI (2)
Christian Tinauer, Anna Damulina, Maximilian Sackl, Martin Soellradl, Reduan Achtibat, Maximilian Dreyer, Frederik Pahde, Sebastian Lapuschkin, Reinhold Schmidt, Stefan Ropele, Wojciech Samek, Christian Langkammer
2024 J jnl
Explainable concept mappings of MRI: Revealing the mechanisms underlying deep learning-based brain disease classification.
CoRR
Christian Tinauer, Anna Damulina, Maximilian Sackl, Martin Soellradl, Reduan Achtibat, Maximilian Dreyer, Frederik Pahde, Sebastian Lapuschkin, Reinhold Schmidt, Stefan Ropele, Wojciech Samek, Christian Langkammer
2024 A* conf
From Hope to Safety: Unlearning Biases of Deep Models via Gradient Penalization in Latent Space.
AAAI
Maximilian Dreyer, Frederik Pahde, Christopher J. Anders, Wojciech Samek, Sebastian Lapuschkin
2024 — conf
PURE: Turning Polysemantic Neurons Into Pure Features by Identifying Relevant Circuits.
XAI4CV
Maximilian Dreyer, Erblina Purelku, Johanna Vielhaben, Wojciech Samek, Sebastian Lapuschkin
2024 J jnl
PURE: Turning Polysemantic Neurons Into Pure Features by Identifying Relevant Circuits.
CoRR
Maximilian Dreyer, Erblina Purelku, Johanna Vielhaben, Wojciech Samek, Sebastian Lapuschkin
2024 J jnl
Pruning By Explaining Revisited: Optimizing Attribution Methods to Prune CNNs and Transformers.
CoRR
Sayed Mohammad Vakilzadeh Hatefi, Maximilian Dreyer, Reduan Achtibat, Thomas Wiegand, Wojciech Samek, Sebastian Lapuschkin
2024 — conf
Pruning by Explaining Revisited: Optimizing Attribution Methods to Prune CNNs and Transformers.
ECCV Workshops (21)
Sayed Mohammad Vakilzadeh Hatefi, Maximilian Dreyer, Reduan Achtibat, Thomas Wiegand, Wojciech Samek, Sebastian Lapuschkin
2024 — conf
Reactive Model Correction: Mitigating Harm to Task-Relevant Features via Conditional Bias Suppression.
CVPR Workshops
Dilyara Bareeva, Maximilian Dreyer, Frederik Pahde, Wojciech Samek, Sebastian Lapuschkin
2024 J jnl
Reactive Model Correction: Mitigating Harm to Task-Relevant Features via Conditional Bias Suppression.
CoRR
Dilyara Bareeva, Maximilian Dreyer, Frederik Pahde, Wojciech Samek, Sebastian Lapuschkin
2024 — conf
Understanding the (Extra-)Ordinary: Validating Deep Model Decisions with Prototypical Concept-based Explanations.
CVPR Workshops
Maximilian Dreyer, Reduan Achtibat, Wojciech Samek, Sebastian Lapuschkin
2023 J jnl
From Hope to Safety: Unlearning Biases of Deep Models by Enforcing the Right Reasons in Latent Space.
CoRR
Maximilian Dreyer, Frederik Pahde, Christopher J. Anders, Wojciech Samek, Sebastian Lapuschkin
2023 J jnl
From attribution maps to human-understandable explanations through Concept Relevance Propagation.
Nat. Mac. Intell.
Reduan Achtibat, Maximilian Dreyer, Ilona Eisenbraun, Sebastian Bosse, Thomas Wiegand, Wojciech Samek, Sebastian Lapuschkin
2023 — conf
Reveal to Revise: An Explainable AI Life Cycle for Iterative Bias Correction of Deep Models.
MICCAI (2)
Frederik Pahde, Maximilian Dreyer, Wojciech Samek, Sebastian Lapuschkin
2023 J jnl
Reveal to Revise: An Explainable AI Life Cycle for Iterative Bias Correction of Deep Models.
CoRR
Frederik Pahde, Maximilian Dreyer, Wojciech Samek, Sebastian Lapuschkin
2023 — conf
Revealing Hidden Context Bias in Segmentation and Object Detection through Concept-specific Explanations.
CVPR Workshops
Maximilian Dreyer, Reduan Achtibat, Thomas Wiegand, Wojciech Samek, Sebastian Lapuschkin
2023 J jnl
Understanding the (Extra-)Ordinary: Validating Deep Model Decisions with Prototypical Concept-based Explanations.
CoRR
Maximilian Dreyer, Reduan Achtibat, Wojciech Samek, Sebastian Lapuschkin
2022 J jnl
From "Where" to "What": Towards Human-Understandable Explanations through Concept Relevance Propagation.
CoRR
Reduan Achtibat, Maximilian Dreyer, Ilona Eisenbraun, Sebastian Bosse, Thomas Wiegand, Wojciech Samek, Sebastian Lapuschkin
2022 J jnl
Revealing Hidden Context Bias in Segmentation and Object Detection through Concept-specific Explanations.
CoRR
Maximilian Dreyer, Reduan Achtibat, Thomas Wiegand, Wojciech Samek, Sebastian Lapuschkin
2021 J jnl
ECQ
CoRR
Daniel Becking, Maximilian Dreyer, Wojciech Samek, Karsten Müller, Sebastian Lapuschkin
2020 — conf
ECQ
xxAI@ICML
Daniel Becking, Maximilian Dreyer, Wojciech Samek, Karsten Müller, Sebastian Lapuschkin