Pascal Frossard

Generating Directed Graphs with Dual Attention and Asymmetric Encoding

Model soups need only one ingredient

Backdoor Unlearning By Linear Task Decomposition

GrapHist: Large-Scale Graph Self-Supervised Learning for Histopathology

Flow based approach for Dynamic Temporal Causal models with non-Gaussian or Heteroscedastic Noises

DeFoG: Discrete Flow Matching for Graph Generation

MEMOIR: Lifelong Model Editing with Minimal Overwrite and Informed Retention for LLMs

Inductive Domain Transfer In Misspecified Simulation-Based Inference

How Compositional Generalization and Creativity Improve as Diffusion Models are Trained

LiNeS: Post-training Layer Scaling Prevents Forgetting and Enhances Model Merging

Pareto Low-Rank Adapters: Efficient Multi-Task Learning with Preferences

DeFoG: Defogging Discrete Flow Matching for Graph Generation