Single-Pass Top-K LRU Victim Selection

Fused Multi-Bank Expert Transfer via Device-Resident Work Lists

Graph-Resident Heterogeneous CPU-GPU Execution Replay

Semantic-Aware LRU Expert Caching in MoE Decode

Bandwidth-Adaptive Miss Partitioning in MoE Decode

Device-Side Dynamic Cache Control via Data-Represented CUDA Graphs

Bandwidth-Adaptive Decode and the q* Execution Policy - Edge-Native Mixture-of-Experts Serving with FreeToken @ University of Michigan - Ann Arbor

Elastic Runtime Cache Reconfiguration and Fast Bootstrap - Edge-Native Mixture-of-Experts Serving with FreeToken @ University of Michigan - Ann Arbor