arXiv AI By Berkcan Kapusuzoglu, Connor Pryor, Sangwoo Cho, Supriyo Chakraborty, Shi-Xiong Zhang, Sambit Sahu, Milind Naphade

When Load-Balancing Goes Too Far: Expert Pruning in Over-Dispersed Mixture-of-Experts Models

Read the original on arXiv AI →

The paper examines how expert pruning—removing low‑importance experts in Mixture‑of‑Experts models—fails when the router is over‑dispersed, a condition caused by aggressive load‑balancing that spreads tokens nearly uniformly across experts. In this regime, traditional importance signals from router probabilities collapse, making perplexity an unreliable predictor of downstream accuracy; for example, the lowest‑perplexity pruning on gpt‑oss‑20B harms mathematical reasoning while the highest‑perplexity pruning preserves it. To address this, the authors introduce Minimax Expert Score Allocation (MESA), a domain‑aware method that iteratively boosts scores for the most affected domain, achieving minimal worst‑case degradation across domains and outperforming baseline pruning strategies on multiple benchmarks while reducing memory usage.

Machine-generated by The Flow from the publisher's headline and feed description — not written or checked by a human. The full article lives at arXiv AI.

arXiv Machine Learning
Aug 27

Ban&Pick: Enhancing Performance and Efficiency of MoE-LLMs via Smarter Routing

The paper introduces Ban&Pick, a post‑training, plug‑and‑play routing strategy for Sparse Mixture‑of‑Experts large language models. It identifies and reinforces a small group of highly influential experts while dynamically pruning redundant ones, leading to accuracy gains across math, code, and reasoning benchmarks. Experiments on DeepSeek and Qwen3 show notable performance improvements and a 1.25× inference speedup without retraining or architectural changes.

By Yuanteng Chen, Peisong Wang, Yuantian Shao, Nanxin Zeng, Chang Xu, Jian Cheng
arXiv Machine Learning
Jun 17

AIMER: Calibration-Free Task-Agnostic MoE Expert Pruning

arXiv:2603. 18492v3 Announce Type: replace Abstract: Mixture-of-Experts (MoE) language models increase parameter capacity without proportional per-token computation, yet deployment still requires storing the full expert pool, making expert pruning important for reducing memory and serving overhead.

By Zongfang Liu, Guangyi Chen, Shengkun Tang, Yifan Shen, Huan Wang, Xin Yuan
arXiv Machine Learning
Jul 15

A Replicate-and-Quantize Strategy for Plug-and-Play Load Balancing of Sparse Mixture-of-Experts LLMs

arXiv:2602. 19938v2 Announce Type: replace Abstract: Sparse Mixture-of-Experts (SMoE) architectures are increasingly used to scale large language models efficiently, delivering strong accuracy under fixed compute budgets.

By Zijie Liu, Jie Peng, Jinhao Duan, Zirui Liu, Kaixiong Zhou, Mingfu Liang, Luke Simon, Xi Liu, Zhaozhuo Xu, Tianlong Chen