We introduce ReaMIL (Reasoning- and Evidence-Aware MIL), a multiple instance learning approach for whole-slide histopathology that adds a light selection head to a strong MIL backbone. The head produces soft per-tile gates and is trained with a budgeted-sufficiency objective: a hinge loss that enforces the true-class probability to be $\geq τ$ using only the kept evidence, under a sparsity budget on the number of selected tiles. The budgeted-sufficiency objective yields small, spatially compact evidence sets without sacrificing baseline performance. Across TCGA-NSCLC (LUAD vs. LUSC), TCGA-BRCA (IDC vs. Others), and PANDA, ReaMIL matches or slightly improves baseline AUC and provides quantitative evidence-efficiency diagnostics. On NSCLC, it attains AUC 0.983 with a mean minimal sufficient K (MSK) $\approx 8.2$ tiles at $τ= 0.90$ and AUKC $\approx 0.864$, showing that class confidence rises sharply and stabilizes once a small set of tiles is kept. The method requires no extra supervision, integrates seamlessly with standard MIL training, and naturally yields slide-level overlays. We report accuracy alongside MSK, AUKC, and contiguity for rigorous evaluation of model behavior on WSIs.

翻译：本文提出ReaMIL（推理与证据感知多示例学习），一种面向全切片组织病理学的多示例学习方法，其在强大的MIL骨干网络上增加了一个轻量级选择头。该选择头生成软化的分块门控信号，并通过预算充足性目标进行训练：该目标采用铰链损失函数，在选定分块数量满足稀疏性预算约束下，强制要求仅使用保留证据时其真实类概率达到 $\geq τ$。预算充足性目标能够在保持基线性能的前提下，生成小而空间紧凑的证据集合。在TCGA-NSCLC（LUAD vs. LUSC）、TCGA-BRCA（IDC vs. Others）和PANDA数据集上的实验表明，ReaMIL在匹配或小幅提升基线AUC的同时，提供了定量的证据效率诊断指标。在NSCLC任务中，该方法在 $τ=0.90$ 时取得AUC 0.983，平均最小充分K值（MSK）$\approx 8.2$ 个分块，AUKC $\approx 0.864$，证明当保留少量分块集合后，类别置信度会急剧上升并趋于稳定。该方法无需额外监督，可与标准MIL训练流程无缝集成，并自然生成切片级可视化热图。我们通过准确率、MSK、AUKC及空间连续性指标，对模型在全切片图像上的行为进行严格评估。