Get started
Author

Jim Berend

Observed name: Berend, Jim
ORCID: 0009-0006-8259-7278
Observed ORCID: 0009-0006-8259-7278
OpenAlex ID: A5092691971
From What to How: Attributing CLIP's Latent Components Reveals Unexpected Semantic Reliance
2025 · Dreyer, Maximilian, Hufe, Lorenz, et al. · arXiv (Cornell University) · 10
PDF
Mechanistic understanding and validation of large AI models with SemanticLens
2025 · Dreyer, Maximilian, Berend, Jim, et al. · arXiv (Cornell University) · 9
PDF
Beyond Scalars: Concept-Based Alignment Analysis in Vision Transformers
2025 · Johanna Vielhaben, Dilyara Bareeva, et al. · neural information processing systems · 6
PDF
Atlas-Alignment: Making Interpretability Transferable Across Language Models
2025 · Puri, Bruno, Berend, Jim, et al. · arXiv (Cornell University) · 5
PDF
Efficient and Flexible Neural Network Training through Layer-wise Feedback Propagation
2023 · Weber, Leander, Berend, Jim, et al. · arXiv (Cornell University) · 3
PDF
Localizing predictive concepts encoded in neural representations via hierarchical testing
2026 · Huryn, Viktoriia, Simnacher, Marco, et al. · arXiv (Cornell University)
PDF
Benchmarking Vector Quantized Auto-Encoders for Multimodal Vision-Language Tokenization of Medical Images
2026 · Emílio Dolgener Cantú, Manasi Acharya, et al. · Companion Publication of the International Conference on Multimodal Interaction (ICMI Companion)
PDF
ResLRP: The Role of Residual Cancellation in Attribution Instability in Vision Transformers
2026 · Berend, Jim, Achtibat, Reduan, et al. · arXiv (Cornell University)
PDF
A review of vision transformer explainability: Methods, evaluation, and future research directions
2026 · Koreuber, Nora, Winklmayr, Claudia, et al. · publish.UP (University of Potsdam)
Playing the network backward: A Game Theoretic Attribution Framework
2026 · Zimmermann, Jakob Paul, Berend, Jim, et al. · arXiv (Cornell University)
PDF
Contrastive Semantic Projection: Faithful Neuron Labeling with Contrastive Examples
2026 · Bouanani, Oussama, Berend, Jim, et al. · arXiv (Cornell University)
PDF
Interpretability-guided Probing for Diagnosing and Mitigating Failure Modes in CLIP - Supplementary Materials
2026 · Dreyer, Maximilian, Hufe, Lorenz, et al. · Mendeley Data
X-SYS: A Reference Architecture for Interactive Explanation Systems
2026 · Labarta, Tobias, Hoang, Nhi, et al. · arXiv (Cornell University)
PDF
Interpretability-Guided Probing for Diagnosing and Mitigating Failure Modes in CLIP
2026 · Wojciech Samek, Maximilian Dreyer, et al. · SSRN Electronic Journal
Works count: 18
Citation count: 57
H-index: 5
I10-index: 2
Year
Topic
Type