AI Paper Atlas

Research map / Architectures and efficiency / Mixture of experts

SpecMD: A Comprehensive Study On Speculative Expert Prefetching

Duc N.M Hoang, Mohammad Samragh, AJAY KUMAR JAISWAL, Minsik Cho

ICML 2026 · Architectures and efficiency › Mixture of experts · approach: Expert routing and specialization · 0 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
replace eviction policy with Least-Stale (ablation)Eliseev '23 (OLMoE-1B-7B) · TTFT, OLMoE-1B-7B, 5% cacheTTFT (ms): 3076.8 -> 2748.1Improved
replace eviction policy with Least-Stale (ablation)Zhu et al. '25 (OLMoE-1B-7B) · TTFT, OLMoE-1B-7B, 5% cacheTTFT (ms): 2257.2 -> 1473.1Improved
replace eviction policy with Least-Stale (ablation)HOBBIT '24 (OLMoE-1B-7B) · TTFT, OLMoE-1B-7B, 5% cacheTTFT (ms): 1590.1 -> 1309.2Improved
replace eviction policy with Least-Stale (ablation)MixtureCache '25 (OLMoE-1B-7B) · TTFT, OLMoE-1B-7B, 5% cacheTTFT (ms): 2664.6 -> 2297.5Improved
replace eviction policy with Least-Stale (ablation)Zhu et al. '25 (Mixtral-8x7B) · TTFT, Mixtral-8x7B, 5% cacheTTFT (ms): 1620.3 -> 1460.8Improved
replace eviction policy with Least-Stale (ablation)Zhu et al. '25 (Phi-3.5-MoE) · TTFT, Phi-3.5-MoE, 5% cacheTTFT (ms): 1905.6 -> 1987.6Hurt
replace eviction policy with Least-Stale (ablation)HOBBIT '24 (Phi-3.5-MoE) · TTFT, Phi-3.5-MoE, 5% cacheTTFT (ms): 3458.6 -> 1933.6Improved
replace eviction policy with Least-Stale (ablation)Eliseev '23 (Qwen1.5-MoE-A2.7B) · TTFT, Qwen1.5-MoE-A2.7B, 5% cacheTTFT (ms): 3724.0 -> 3433.4Improved
use Least-Stale instead of LRU eviction (ablation)SpecMD · collision misses, 5% cachecollision miss rate: 1.6-1.9% vs 4.5-12.6%Improved
use Least-Stale instead of Score-Based eviction (ablation)SpecMD · collision misses, 5% cachecollision miss rate: 1.6-1.9% vs 42.6-60.9%Improved
use score-based prefetching instead of top-k (ablation)SpecMD with Least-Stale · hit rate / sync overhead, 5% cachehit rateImproved
use substitution of missing experts with cached experts (ablation)SpecMD · GSM8K, 5% cacheexact match: 0.691 -> 0.073Hurt

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy