logo
today local_bar

Arthur Douillard

Researcher, DeepMind

2 papers at NeurIPS 2025

Homepage· OpenReview· Semantic Scholar· Google Scholar

Poster Session 3

1 paper
Thursday, December 4, 2025 · 11:00 AM → 2:00 PM
Exhibit Hall C,D,E
Communication-Efficient Language Model Training Scales Reliably and Robustly: Scaling Laws for DiLoCo
#811 Spotlight · Zachary Charles, Gabriel Teston, Lucio M. Dery, J Keith Rush, Nova Fallen, Zachary Garrett, Arthur Szlam, Arthur Douillard

We show that DiLoCo, a method for communication-efficient language model training, exhibits reliable scaling law behavior.

Poster Session 6

1 paper
Friday, December 5, 2025 · 4:30 PM → 7:30 PM
Exhibit Hall C,D,E
Understanding outer learning rates in Local SGD
#801 · Ahmed Khaled, Satyen Kale, Arthur Douillard, Chi Jin, Rob Fergus, Manzil Zaheer