AI Paper Atlas

Research map / Architectures and efficiency / Mixture of experts

RaGEP: Rank-aware Geometric Expert Pruning for Mixture-of-Experts Language Models

Wentao Hu, Zeyu Zhu, Mingkuan Zhao, Zhenhua An, Yanbo Zhai, Shanhong yu, Huilin Zhou, Xin Lai, Xiaoyan Zhu, Jiayin Wang

ICML 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
Add rank-aware budget allocation (ablation)EEP · Qwen3-30B-A3B, retain 50%average accuracy: 61.60 -> 61.96Improved
Add spectral-salience pruning (ablation)EEP · Qwen3-30B-A3B, retain 50%average accuracy: 61.60 -> 62.18Improved
Combine rank-aware budget allocation and SSP (ablation)EEP · Qwen3-30B-A3B, retain 50%average accuracy: 61.60 -> 63.12Improved
Add rank-aware budget allocation (ablation)EEP · Qwen3-30B-A3B, retain 75%average accuracy: 66.96 -> 67.53Improved
Add spectral-salience pruning (ablation)EEP · Qwen3-30B-A3B, retain 75%average accuracy: 66.96 -> 67.30Improved
Combine both components (ablation)EEP · Qwen3-30B-A3B, retain 75%average accuracy: 66.96 -> 67.72Improved
Add rank-aware budget allocation (MMLU only) (ablation)EEP · Qwen3-30B-A3B MMLU, retain 50%accuracy: 55.67 -> 53.12Hurt
Add spectral-salience pruning (MMLU only) (ablation)EEP · Qwen3-30B-A3B MMLU, retain 50%accuracy: 55.67 -> 50.80Hurt
Raise balancing factor beta from 0.0 to 0.8 (ablation)RaGEP · DeepSeek-V2-Liteaverage accuracyImproved
Raise beta from 0.8 to 1.0 (pure salience) (ablation)RaGEP · DeepSeek-V2-Liteaverage accuracyHurt
Replace EEP/MoP selection with RaGEPEEP · DeepSeek-V2-Lite, 50% pruningaverage accuracy: 45.70 -> 47.33Improved
Replace EEP with RaGEPEEP · Mixtral 8x7B, 50% pruningaverage accuracy: 62.86 -> 63.83Improved

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