YOLO-Master：基于专用Transformer的MOE加速增强实时检测
YOLO-Master: MOE-Accelerated with Specialized Transformers for Enhanced Real-time Detection
将稀疏混合专家（MoE）与动态路由引入YOLO框架，实现按场景复杂度自适应分配计算资源，为实时目标检测提供可复用的高效计算范式。
原文依据 we propose YOLO-Master, a novel YOLO-like framework that introduces image-level adaptive computation for RTOD.
摘要译文
由模型翻译 · 术语统一 · 请对照原文核对关键结论现有的实时目标检测（RTOD）方法通常采用类YOLO架构，因其在精度与速度之间取得了良好的权衡。然而，这些模型依赖于静态稠密计算，对所有输入进行统一处理，导致表征能力和计算资源（如对简单场景过度分配、对复杂场景分配不足）的错配。这种错配不仅造成计算冗余，还导致检测性能次优。为克服这一局限，我们提出了YOLO-Master，一种新颖的类YOLO框架，为RTOD引入图像级自适应计算。这是通过高效稀疏混合专家（ES-MoE）模块实现的，该模块根据每个输入的场景复杂度动态分配计算资源。其核心是一个轻量级动态路由网络，通过多样性增强目标在训练期间引导专家专业化，鼓励专家之间形成互补的专业知识。此外，路由网络自适应地学习仅激活最相关的专家，从而在推理时提升检测性能的同时最小化计算开销。在五个大规模基准上的综合实验证明了YOLO-Master的优越性。在MS COCO上，我们的模型以1.62毫秒的延迟达到42.4%的AP，相比YOLOv13-N提升了+0.8%的mAP，推理速度加快17.8%。值得注意的是，在具有挑战性的密集场景中增益最为显著，同时模型在典型输入上保持效率并维持实时推理速度。Github: https://github.com/Tencent/YOLO-Master/
英文原文摘要 1624 字符
Existing Real-Time Object Detection (RTOD) methods commonly adopt YOLO-like architectures for their favorable trade-off between accuracy and speed. However, these models rely on static dense computation that applies uniform processing to all inputs, misallocating representational capacity and computational resources such as over-allocating on trivial scenes while under-serving complex ones. This mismatch results in both computational redundancy and suboptimal detection performance.To overcome this limitation, we propose YOLO-Master, a novel YOLO-like framework that introduces image-level adaptive computation for RTOD. This is achieved through an Efficient Sparse Mixture-of-Experts (ES-MoE) block that dynamically allocates computational resources to each input according to its scene complexity. At its core, a lightweight dynamic routing network guides expert specialization during training through a diversity enhancing objective, encouraging complementary expertise among experts. Additionally, the routing network adaptively learns to activate only the most relevant experts, thereby improving detection performance while minimizing computational overhead during inference.Comprehensive experiments on five large-scale benchmarks demonstrate the superiority of YOLO-Master. On MS COCO, our model achieves 42.4% AP with 1.62ms latency, outperforming YOLOv13-N by +0.8% mAP and 17.8% faster inference. Notably, the gains are most pronounced on challenging dense scenes, while the model preserves efficiency on typical inputs and maintains real-time inference speed. Github: https://github.com/Tencent/YOLO-Master/