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. We propose Hierarchical Instance-Conditioned Mixture-of-Experts (HI-MoE), a DETR-style detection architecture that performs routing in two stages: a lightweight scene router first selects a scene-consistent expert subset, and an instance router then assigns each object query to a small number of experts within that subset. This design aims to preserve sparse computation while better matching the heterogeneous, instance-centric structure of detection. In the current draft, experiments are concentrated on COCO with preliminary specialization analysis on LVIS. Under these settings, HI-MoE improves over a dense DINO baseline and over simpler token-level or instance-only routing variants, with especially strong gains on small objects. We also provide an initial visualization of expert specialization patterns. We present the method, ablations, and current limitations in a form intended to support further experimental validation.

翻译：混合专家（MoE）架构通过仅为每个输入激活部分模型参数，实现了条件计算。尽管稀疏路由在语言模型中效果显著，并在视觉领域也展现出潜力，但大多数视觉MoE方法在图像或补丁级别运行。这种粒度与目标检测任务契合度较低，因为在该任务中，推理的基本单元是与候选实例相对应的对象查询。我们提出分层实例条件混合专家（HI-MoE），这是一种DETR风格的检测架构，分两个阶段执行路由：轻量级场景路由器首先选择与场景一致的专家子集，然后实例路由器将每个对象查询分配给该子集中的少量专家。该设计旨在保持稀疏计算的同时，更好地匹配目标检测中异构、以实例为中心的结构特征。在当前版本中，实验集中在COCO数据集上，并在LVIS数据集上进行了初步的专业化分析。在这些设置下，HI-MoE相较于密集的DINO基线以及更简单的令牌级或实例级路由变体均有所改进，尤其在小目标上表现突出。我们还提供了专家专业化模式的初步可视化。我们以支持进一步实验验证的形式，介绍了该方法、消融实验及当前局限性。