SlimWise: Decoupling Expert Pruning Across Prefill and Decode for Efficient MoE Serving

arXiv:2609.34117 · cs.LG, cs.AI · Submitted 2026-09-28 · Read on arXiv

cs.LG, cs.AI

Submitted: 2026-09-28

Updated: 2026-09-28

Code: https://github.com/huggingface/math-verify

License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/

Terminology

• Decode — 1× in this paper · explained in 2 episodes
• Decoupling — 1× in this paper · explained in 2 episodes
• Pruning — 1× in this paper · explained in 4 episodes

Sources

• NVIDIA Nemotron Nano 2: An Accurate and Efficient Hybrid Mamba-Transformer Reasoning Model
• Retraining-Free Merging of Sparse MoE via Hierarchical Clustering
• ELDR: Expert-Locality-Aware Decode Routing for PD-Disaggregated MoE Serving
• Training Verifiers to Solve Math Word Problems
• Gemma 4 Technical Report
• Lynx: Enabling Efficient MoE Inference through Dynamic Batch-Aware Expert Selection
• LoRA: Low-Rank Adaptation of Large Language Models
• Finding Fantastic Experts in MoEs: A Unified Study for Expert Dropping Strategies and Observations
• Mixtral of Experts
• Pruning and Distilling Mixture-of-Experts into Dense Language Models
• DeepSeek-V3 Technical Report
• Quantized Reasoning Models Think They Need to Think Longer, but They Do Not
• Opportunistic Expert Activation: Batch-Aware Expert Routing for Faster Decode Without Retraining
• SlimQwen: Exploring the Pruning and Distillation in Large MoE Model Pre-training
• XShare: Collaborative in-Batch Expert Sharing for Faster MoE Inference
• ExFold: Unified Expert Folding for Training-Free MoE Prefill-Decode Acceleration
• MoE-Pruner: Pruning Mixture-of-Experts Large Language Model using the Hints from Its Router
• Qwen3 Technical Report
• Janus: Disaggregating Attention and Experts for Scalable MoE Inference

Related papers

• Polynomial-Augmented Neural Networks (PANNs) with Weak Orthogonality Constraints for Enhanced Function and PDE Approximation
• AIRL-S: Unifying Reinforcement Learning and Search-Based Test-Time Scaling via Adversarial Inverse Reinforcement Learning
• Transformers as Bayesian In-Context Experimenters: Smoothness-Adaptive Efficient ATE Estimation
• Convergence issues in Relational Concept Analysis based on AOC-posets
• Beliefs Beyond Posteriors: The Predictive Cost of ELBO Binding
• Understanding Diffusion Models via Ratio-Based Function Approximation with SignReLU Networks