arXiv:2607. 11586v1 Announce Type: new Abstract: Mixture-of-Experts (MoE) large language models (LLM) activate only a small number of experts during inference, but token routing introduces persistent expert hotness skew: a small set of hot experts continuously receives most tokens, while the remaining experts are lightly loaded.
By Yongqin Zhang
arXiv:2607. 08782v1 Announce Type: cross Abstract: Expert parallelism has become the prevailing paradigm to serve Mixture-of-Experts (MoE) models.
By Qianli Liu, Kaibin Guo, Zicong Hong, Peng Li, Fahao Chen, Haodong Wang, Jian Lin, Song Guo
arXiv:2609.36222v1 Announce Type: new
Abstract: Large language models are increasingly expensive to serve. In large-scale serving systems, autoregressive decoding is often bottlenecked by transferrin...
By Ali Abbasi, Justin Shi, Soheil Kolouri
FluxMoE introduces an expert paging system that decouples Mixture-of-Experts (MoE) model experts from permanent GPU residency, allowing dynamic adaptation to available memory. By combining PagedTensor, a bandwidth‑balanced memory hierarchy, and a budget‑aware residency planner, FluxMoE streams expert weights on demand while keeping computations on GPUs. Experiments on GLM‑4.5 and Mixtral‑8×7B‑Instruct show significant throughput gains and reduced time‑per‑output‑token compared to existing inference engines, without compromising model quality.
By Qingxiu Liu, Yongchao He, Runhan Jiang, Zion Wang, Bohan Zhao, Mi Zhang, Patrick P. C. Lee
arXiv:2608. 07964v1 Announce Type: cross Abstract: Load Balancing has emerged as a critical problem in expert-parallel distributed inference of Mixture-of-Experts (MoE) models.
By Yize Wu, Ke Gao, Ling Li, Yanjun Wu
arXiv:2607. 19539v1 Announce Type: cross Abstract: Mixture-of-Experts (MoE) architectures increase model capacity without proportionally increasing computation cost and have become a key building block for scaling large language models (LLMs) to trillion-parameter regimes.
By Minyu Cui, Anna Wingkvist, Morgan Ericsson