Can a lightweight, severity-aware adversarial detection filter (e.g., based on embedding cosine distance) impr

State-of-the-art object detection networks depend on region proposal algorithms to hypothesize object locations. Advances like SPPnet [1] and Fast R-CNN [2] have reduced the running time of these detection networks, exposing region proposal computation as a bottleneck. In this work, we introduce a Region Proposal Network (RPN) that shares full-image convolutional features with the detection network, thus enabling nearly cost-free region proposals. An RPN is a fully convolutional network that simultaneously predicts object bounds and objectness scores at each position. The RPN is trained end-to

Research goal: Can a lightweight, severity-aware adversarial detection filter (e.g., based on embedding cosine distance) improve inference throughput by selectively skipping retrieval or generation steps in RAG pipelines, and what is the throughput-accuracy trade-off on MultiHopQA under adversarial perturbations?

Autonomous synthesis report generated by SOVEREIGN Research Kernel. Tribunal consensus score: 8.7/10.