AI Paper Atlas

Research map / Architectures and efficiency / Mixture of experts

Steering MoE LLMs via Expert (De)Activation

Mohsen Fayyaz, Ali Modarressi, Hanieh Deilamsalehy, Franck Dernoncourt, Ryan A. Rossi, Trung Bui, Hinrich Schuetze, Nanyun Peng

ICLR 2026 · Architectures and efficiency › Mixture of experts · approach: Expert routing and specialization · 20 citations (OpenAlex)

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 SteerMoE unsafe-expert steering to the AIM jailbreakGPT-OSS-120B · AdvBench (50 prompts)safe response rate: 100% -> 0%Improved
add SteerMoE unsafe-expert steering to FFAGPT-OSS-120B · AdvBench (50 prompts)safe response rate: 100% -> 18%Improved
add SteerMoE to AIMOLMoE · AdvBench (50 prompts)safe response rate: 100% -> 36%Improved
add SteerMoE to AIMQwen3-30B-A3B · AdvBench (50 prompts)safe response rate: 2% -> 2%No clear effect
add SteerMoE to FFAQwen3-30B-A3B · AdvBench (50 prompts)safe response rate: 48% -> 48%No clear effect
use risk difference instead of log-odds ratio to score experts (ablation)SteerMoE · MQuAKE faithfulness, Qwen3faithfulness accuracy: 78% -> 97%Improved
use log-odds ratio scoring with more experts deactivated (ablation)SteerMoE · MQuAKE faithfulness, Qwen3faithfulness accuracy: 84% -> 78%Hurt
deactivate more risk-difference experts (100 to 500) (ablation)SteerMoE · MQuAKE faithfulness, Qwen3faithfulness accuracy: 86% -> 97%Improved
select experts by risk difference instead of randomly or bottom-k (ablation)SteerMoE · AdvBench GPT-OSS-120Bsafe response rate: 100% -> 0%Improved
apply SteerMoE detected on English to other languagesSteerMoE + AIM · MultiJail, GPT-OSS-120Bsafe response rate: English 94.3%, Italian 90.2%, Thai 87.9% -> 9.5%, 11.4%, 7.3% with AIMImproved
steer toward faithful expertsQwen3 and other MoE LLMs · FaithEval, CF-TriviaQA, MQuAKEfaithfulness accuracy: up to +27%Improved
steer toward safe expertsMoE LLMs · safety red-teaming benchmarkssafe response rate: up to +20%Improved

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