Björn Deiseroth

u-$μ$P: The Unit-Scaled Maximal Update Parametrization

arXiv 2024

T-FREE: Subword Tokenizer-Free Generative LLMs via Sparse Representations for Memory-Efficient Embeddings

arXiv 2024

SCAR: Sparse Conditioned Autoencoders for Concept Detection and Steering in LLMs

arXiv 2024

AtMan: Understanding Transformer Predictions Through Memory Efficient Attention Manipulation

NeurIPS 2023 11

Safe Latent Diffusion: Mitigating Inappropriate Degeneration in Diffusion Models

CVPR 2023 1

ILLUME: Rationalizing Vision-Language Models through Human Interactions

arXiv 2022

Speaking Multiple Languages Affects the Moral Bias of Language Models

arXiv 2022

Kristian Kersting

Patrick Schramowski

Manuel Brack

Samuel Weinbach

Alexander Fraser

Andres Felipe Cruz-Salinas

Carlo Luschi

Charlie Blake

Constantin A. Rothkopf

Constantin Eichenberg