Rethinking MoE Load Balancing as Feedback Control
cs.LG
Submitted: 2026-11-06
Updated: 2026-11-06
Terminology
Sources
- Learning Factored Representations in a Deep Mixture of Experts
- A Theoretical Framework for Auxiliary-Loss-Free Load Balancing of Sparse Mixture-of-Experts in Large-Scale AI Models
- Learning Sparse Latent Predictive Foundation Model for Multimodal Neuroimaging
- Mixture of Experts for Low-Resource LLMs
- Mellum2 Technical Report
- Innovator: Scientific Continued Pretraining with Fine-grained MoE Upcycling
- DeepSeek-V3 Technical Report
- LongCat-Flash Technical Report
- Auxiliary-Loss-Free Load Balancing Strategy for Mixture-of-Experts
- DeepSeek-Coder-V2: Breaking the Barrier of Closed-Source Models in Code Intelligence
Related papers
- A Physics-Regulated Neural Framework for Learning 3D Grain Growth Dynamics
- FlexMoE: Learning One-for-All Multi-Budget Expert Pruning for Flexible MoE Deployment
- OptiMind: Improving Small Language Models for Optimization Modeling through Formulation Guidance and Multi-Turn Learning
- Staying in the Sweet Spot: Boost Reasoning Evolution via Responsive Rollout Difficulty Manipulation
- Examining the Impact of Data and Training Methods on Emergent and Subliminal Misalignment
- Rethinking Black-Box On-Policy Distillation: A Rubric-based Analysis