목록LLM Inference (1)

Engineering insight

[arXiv-2026] FlashMoE: Reducing SSD I/O Bottlenecks via ML-Based Cache Replacement for Mixture-of-Experts Inference on Edge Devices

FlashMoE: Reducing SSD I/O Bottlenecks via ML-Based Cache Replacement for Mixture-of-Experts Inference on Edge DevicesarXiv:2601.17063v1 (2026) · Sangyeob Kim et al. · HTML: https://arxiv.org/html/2601.17063v1 FlashMoE: Reducing SSD I/O Bottlenecks via ML-Based Cache Replacement for Mixture-of-Experts Inference on Edge DevicesFlashMoE: Reducing SSD I/O Bottlenecks via ML-Based Cache Replacement ..
AI/AI Paper review 2026. 5. 5. 23:39
이전 Prev 1 Next 다음