QG-MIL：一种用于医学影像中领域无关多实例学习的门控Transformer聚合器
QG-MIL: A Gated Transformer Aggregator for Domain-Agnostic Multiple Instance Learning in Medical Imaging
AI总结：
提出QG-MIL门控Transformer聚合器，通过RMSNorm预归一化、逐头QK归一化、细粒度注意力输出门控和SwiGLU前馈模块，解决注意力集中问题，在六个基准上平均提升+6.1个宏F1分数。
摘要：
医学影像中基于注意力的多实例学习聚合器容易出现注意力集中，导致预测过于自信且不稳定。我们引入QG-MIL，一种门控Transformer聚合器，通过四个协同架构组件解决这一问题：基于RMSNorm的预归一化、逐头QK归一化、细粒度注意力输出门控和SwiGLU风格的前馈模块。这些设计选择共同稳定了训练，并将注意力更均匀地分布在实例上，无需辅助损失、掩码或多阶段正则化。我们在涵盖全切片病理学和细胞级血液学的六个基准上评估了QG-MIL，覆盖两种根本不同的MIL尺度。性能最佳的QG-MIL变体在所有六个基准上均优于领先的基线，平均提升+6.1个宏F1分数。注意力覆盖图和注意力质量分析证实了更分布的实例权重。消融研究表明，虽然单个组件在特定数据集上可以匹配完整模型，但与所选基线相比，QG-MIL设计提供了最一致的跨域性能和最紧凑的方差。我们发布了一个可配置的实现以支持可重复性，网址为：this https URL
英文摘要：
Attention-based Multiple Instance Learning aggregators in medical imaging are prone to attention concentration, producing overconfident and unstable predictions. We introduce QG-MIL, a gated transformer aggregator that addresses this through four synergistic architectural components: RMSNorm-based pre-normalization, per-head QK normalization, fine-grained attention output gating, and SwiGLU-style feed-forward modules. Together, these design choices stabilize training and distribute attention more uniformly across instances without auxiliary losses, masking, or multi-stage regularization. We evaluate QG-MIL across six benchmarks spanning whole-slide pathology and cell-level hematology, covering two fundamentally different MIL scales. The best-performing QG-MIL variants outperform leading baselines on all six benchmarks, with an average improvement of +6.1 mean macro F1 points. Attention overlays and attention mass analysis confirm more distributed instance weighting. Ablation studies show that while individual components can match the full model on specific datasets, the QG-MIL design provides the most consistent cross-domain performance and tightest variance when compared to selected baselines. We release a configurable implementation to support reproducibility at: https://github.com/unica-visual-intelligence-lab/QG-MIL
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
