AI Paper Atlas

Research map / Architectures and efficiency / Mixture of experts

Understanding Routing Mechanism in Mixture-of-Experts Language Models

Wengang Li, Lingqi Zhang, Toshio Endo, Mohamed Wahib

ICLR 2026 · Architectures and efficiency › Mixture of experts · approach: Expert routing and specialization

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
zero the routing scores assigned by expert M1E9 (ablation)OLMoE · OLMoE routing score variancevariance stripe at sending layer M1Hurt
zero the routing scores assigned by expert M4E14 (ablation)OLMoE · OLMoE routing score variancevariance stripe at sending layer M4Hurt
zero the scores assigned by M1E18 (control expert) (ablation)OLMoE · OLMoE routing score variancevariance stripesNo clear effect
pad output of M1E9 with zeros (ablation)OLMoE · OLMoE routing score variancevariance stripesHurt
pad output of M4E14 with zeros (ablation)OLMoE · OLMoE routing score variancevariance stripesMixed
pad output of M1E18 with zeros (ablation)OLMoE · OLMoE routing score variancevariance stripesNo clear effect

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy