HI-MoE: Hierarchical Instance-Conditioned Mixture-of-Experts for Object Detection

ArXi:2604.04908v1 Announce Type: new Mixture-of-Experts (MoE) architectures enable conditional computation by activating only a subset of model parameters for each input. Although sparse routing has been highly effective in language models and has also shown promise in vision, most vision MoE methods operate at the image or patch level. This granularity is poorly aligned with object detection, where the fundamental unit of reasoning is an object query corresponding to a candidate instance.