RF-DETR is a real-time transformer architecture for object detection, instance segmentation, and keypoint detection (preview) developed by Roboflow. Built on a DINOv2 vision transformer backbone, ...
Abstract: The adoption of artificial intelligence models (e.g., DNN models) in internet of things has boosted computing demands in edge computing. Frequent model retraining, necessitated by concept ...
LoGo dynamically selects and merges multiple LoRA adapters at inference time based on each input instance, rather than using a single fixed adapter. For each input, LoGo projects the hidden state ...
Abstract: The temporal decay of trust data in dynamic edge computing environments leads to inaccurate evaluation, and the recommended trust values from heterogeneous nodes are affected by subjective ...
Some results have been hidden because they may be inaccessible to you
Show inaccessible results