Martin Jaggi

mjaggi's activity

• All
• Models
• Datasets
• Spaces
• Papers
• Collections
• Community
• Posts
• Upvotes
• Likes
• Articles

authored a paper about 19 hours ago

FineWeb2: One Pipeline to Scale Them All -- Adapting Pre-Training Data Processing to Every Language

Paper • 2506.20920 • Published 2 days ago • 23

authored 5 papers over 1 year ago

Model Fusion via Optimal Transport

Paper • 1910.05653 • Published Oct 12, 2019 • 1

Evaluating the Search Phase of Neural Architecture Search

Paper • 1902.08142 • Published Feb 21, 2019

Second-order optimization with lazy Hessians

Paper • 2212.00781 • Published Dec 1, 2022

Landmark Attention: Random-Access Infinite Context Length for Transformers

Paper • 2305.16300 • Published May 25, 2023

Faster Causal Attention Over Large Sequences Through Sparse Flash Attention

Paper • 2306.01160 • Published Jun 1, 2023 • 1