Georgios Kaissis

A Structured, Tagged, and Localized Visual Question Answering Dataset with Full Sentence Answers and Scene Graphs for Chest X-ray Images

Redirection for Erasing Memory (REM): Towards a universal unlearning method for corrupted data

Efficient numeracy in language models through single-token number embeddings

Unifying Re-Identification, Attribute Inference, and Data Reconstruction Risks in Differential Privacy

Exploring the limits of strong membership inference attacks on large language models

Gradient-Weight Alignment as a Train-Time Proxy for Generalization in Classification Tasks

Laplace Sample Information: Data Informativeness Through a Bayesian Lens

Scaling Laws for Differentially Private Language Models

Improved Localized Machine Unlearning Through the Lens of Memorization

Structured Joint Aleatoric and Epistemic Uncertainty for High Dimensional Output Spaces

On Gradient-Weight Alignment