G. Farnadi

Mila – Quebec AI Institute

Mila

Papers on Lattice

Total citations

Topics

h-index

Publication activitypapers/week, last 8 weeks

Research focus

Architecture Design (Transformers, SSMs, MoE) (1)Natural Language Processing (1)Training Efficiency & Optimization (1)

Frequent co-authors

Aditi Khandelwal (1)Aditi Khandelwal (1)Marius Mosbach (1)Marius Mosbach (1)

Papers (1)

May 28, 2026

Microsoft Research3d ago·also Mila, Edinburgh

Leveraging Routing Dynamics in Mixture-of-Experts Models for Efficient Language Adaptation

Language specialization in multilingual MoEs happens mostly in the final layers, suggesting a surprisingly simple recipe for parameter-efficient adaptation.

Aditi Khandelwal, Aditi Khandelwal, Marius Mosbach +7

Architecture Design (Transformers, SSMs, MoE)Natural Language Processing Training Efficiency & Optimization

Search

G. Farnadi

Publication activitypapers/week, last 8 weeks

Research focus

Frequent co-authors

Papers (1)