Measurement-Driven Sub-Network Selection for On-Premise Retrieval-Augmented Factory Agents
Deployment is treated as a post-adaptation selection problem, committing one sub-network per device on judged answer quality and measured on-device throughput under a configurable general-capability floor and memory budget; rules that optimize size, speed, or quality alone each give up capability or throughput.
V. Rizeakos, Georgios Paisios, Alexandros Machairas et al.
· 0 citations