Specter: Fast MoE Inference via Quantization-Enabled Multi-Step Speculative Offloading
Paper • 2511.14102 • Published • 19
Liu
beannn
·
AI & ML interests
None yet
Recent Activity
upvoted a paper about 19 hours ago
Specter: Fast MoE Inference via Quantization-Enabled Multi-Step Speculative Offloading submitted a paper about 19 hours ago
Specter: Fast MoE Inference via Quantization-Enabled Multi-Step Speculative Offloading upvoted a paper 23 days ago
YuE2: Unifying Symbolic and Audio Music Generation at Frontier QualityOrganizations
None yet