Arthur Szlam

8

Papers

12

Total Citations

Papers (8)

Communication-Efficient Language Model Training Scales Reliably and Robustly: Scaling Laws for DiLoCo

Hard Mixtures of Experts for Large Scale Weakly Supervised Vision

Low-Shot Learning With Large-Scale Diffusion

Order-Aware Generative Modeling Using the 3D-Craft Dataset

Recurrent Orthogonal Networks and Long-Memory Tasks

Optimizing the Latent Space of Generative Networks

Modeling Others using Oneself in Multi-Agent Reinforcement Learning

Composable Planning with Attributes