expert routing

LoRACoE: Improving Large Language Model via Composition-based LoRA Expert EMNLP 2025 Maximum Score Routing For Mixture-of-Experts ACL 2025 MergeME: Model Merging Techniques for Homogeneous and Heterogeneous MoEs NAACL 2025 MidPO: Dual Preference Optimization for Safety and Helpfulness in Large Language Models via a Mixture of Experts Framework EMNLP 2025 Mixture of Parrots: Experts improve memorization more than reasoning ICLR 2025 MoE++: Accelerating Mixture-of-Experts Methods with Zero-Computation Experts ICLR 2025 MultiPL-MoE: Multi-Programming-Lingual Extension of Large Language Models through Hybrid Mixture-of-Experts EMNLP 2025 No Need to Talk: Asynchronous Mixture of Language Models ICLR 2025 OLMoE: Open Mixture-of-Experts Language Models ICLR 2025 Probing Semantic Routing in Large Mixture-of-Expert Models EMNLP 2025 Self-MoE: Towards Compositional Large Language Models with Self-Specialized Experts ICLR 2025 TC-MoE: Augmenting Mixture of Experts with Ternary Expert Choice ICLR 2025 THOR-MoE: Hierarchical Task-Guided and Context-Responsive Routing for Neural Machine Translation ACL 2025 Understanding and Leveraging the Expert Specialization of Context Faithfulness in Mixture-of-Experts LLMs EMNLP 2025 Read-ME: Refactorizing LLMs as Router-Decoupled Mixture of Experts with System Co-Design NeurIPS 2024 HyperMoE: Towards Better Mixture of Experts via Transferring Among Experts ACL 2024 LLaMA-MoE: Building Mixture-of-Experts from LLaMA with Continual Pre-Training EMNLP 2024 LocMoE: A Low-overhead MoE for Large Language Model Training IJCAI 2024 Mixture of Diverse Size Experts EMNLP 2024 MomentumSMoE: Integrating Momentum into Sparse Mixture of Experts NeurIPS 2024 MoVA: Adapting Mixture of Vision Experts to Multimodal Context NeurIPS 2024 Routing to the Expert: Efficient Reward-guided Ensemble of Large Language Models NAACL 2024 Turn Waste into Worth: Rectifying Top-k Router of MoE EMNLP 2024 On the Benefits of Learning to Route in Mixture-of-Experts Models EMNLP 2023 Patch-level Routing in Mixture-of-Experts is Provably Sample-efficient for Convolutional Neural Networks ICML 2023