SpecMoE: A Fast and Efficient Mixture-of-Experts Inference via Self-Assisted Speculative Decoding
The Mixture-of-Experts (MoE) architecture has emerged as a promising approach to mitigate the rising computational costs of large language models (LLMs) by selectively activating parameters. However, its high memory requirements and sub-optimal parameter efficiency pose significant challenges for efficient deployment. Although CPU-offloaded MoE inference systems have been proposed in the literature, they offer limited efficiency, particularly for large batch sizes. In this work, we propose SpecMoE, a memory-efficient MoE inference system based on our self-assisted speculative decoding algorithm. SpecMoE demonstrates the effectiveness of applying speculative decoding to MoE inference witho...
| Authors | Jehyeon Bang, Eunyeong Cho, Ranggi Hwang et al. |
|---|---|
| Primary category | cs.AI |
| Published | 11 April 2026 |
| Last updated at source | 11 April 2026 |
Silytics shows public metadata from arXiv and links to the original. Information may be incomplete or out of date. Last refreshed 17 October 2026. Licenses and terms of the original apply to the work itself. Not financial, legal or professional advice.