mlsanityData to 14 Oct 2026

Statistical Testing for Multiple Instance Learning via Selective Inference with Applications to Computational Pathology

arXiv 2609.31712 ↗First posted 21 Sep 2026cs.CV cs.LG stat.MLcited by 0

Not on the map. Why?

Abstract (arXiv)

Multiple instance learning (MIL) is widely used in computational pathology because it enables weakly supervised analysis of whole-slide images (WSIs) without requiring patch-level annotations. In attention-based MIL, instances with high attention scores are often interpreted as diagnostically important regions and used as visual explanations. However, attention scores alone cannot determine whether selected high-attention instances are significantly different from normal instances, limiting the reliability of attention-based explanations. In this paper, we formulate the evaluation of high-attention instances as a statistical hypothesis testing problem. Specifically, we assess whether a selected high-attention instance significantly deviates from a representative normal reference instance selected based on feature similarity. A major challenge is that both the target instance and the reference instance are selected through data-dependent procedures, rendering standard hypothesis testing invalid. To address this issue, we introduce a selective inference (SI) framework that explicitly accounts for the selection events induced by attention-based instance selection and adaptive reference selection, thereby enabling the computation of valid selective p-values conditional on these events. Experiments demonstrate Type-I error control on synthetic and MNIST-based data and practical applicability to pathological WSIs, with higher statistical power than the conventional over-conditioning approach.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Builds on 37 of its 70 references that are in the map. First posted on arXiv 21 Sep 2026.

Builds on

37 of 70 references.

References · 37

  1. 2017Attention Is All You Needcited by 60,361
  2. 2018Attention-based Deep Multiple Instance Learningcited by 1,095
  3. 2019Attention is not Explanationcited by 929
  4. 2020Data Efficient and Weakly Supervised Computational Pathology on Whole Slide Imagescited by 745
  5. 2019Attention is not not Explanationcited by 636
  6. 2021TransMIL: Transformer based Correlated Multiple Instance Learning for Whole Slide Image Classificationcited by 555
  7. 2020Dual-stream Multiple Instance Learning Network for Whole Slide Image Classification with Self-supervised Contrastive Learningcited by 478
  8. 2019Is Attention Interpretable?cited by 389
  9. 2022Scaling Vision Transformers to Gigapixel Images via Hierarchical Self-Supervised Learningcited by 359
  10. 2022DTFD-MIL: Double-Tier Feature Distillation Multiple Instance Learning for Histopathology Whole Slide Image Classificationcited by 256
  11. 2024Multimodal Whole Slide Foundation Model for Pathologycited by 162
  12. 2016Revisiting Multiple Instance Neural Networkscited by 119
25 more
  1. 2013Exact post-selection inference, with application to the lassocited by 91
  2. 2020Multi-scale Domain-adversarial Multiple-instance CNN for Cancer Subtype Classification with Unannotated Histopathological Imagescited by 81
  3. 2014Optimal Inference After Model Selectioncited by 67
  4. 2017Thoracic Disease Identification and Localization with Limited Supervisioncited by 55
  5. 2014Exact Post-Selection Inference for Sequential Regression Procedurescited by 48
  6. 2020Selective Inference for Hierarchical Clusteringcited by 39
  7. 2015Classifying and Segmenting Microscopy Images Using Convolutional Multiple Instance Learningcited by 38
  8. 2017Valid Inference Corrected for Outlier Removalcited by 26
  9. 2020Computing Valid p-value for Optimal Changepoint by Selective Inference using Dynamic Programmingcited by 25
  10. 2021More Powerful Conditional Selective Inference for Generalized Lasso by Parametric Programmingcited by 25
  11. 2022Selective inference for k-means clusteringcited by 22
  12. 2020Quantifying Statistical Significance of Neural Network-based Image Segmentation by Selective Inferencecited by 21
  13. 2023Valid P-Value for Deep Learning-Driven Salient Regioncited by 17
  14. 2016Post Selection Inference with Kernelscited by 13
  15. 2021Label Cleaning Multiple Instance Learning: Refining Coarse Annotations on Single Whole-Slide Imagescited by 13
  16. 2025Label-free Concept Based Multiple Instance Learning for Gigapixel Histopathologycited by 8
  17. 2025si4onnx: A Python package for Selective Inference in Deep Learning Modelscited by 5
  18. 2024Inference on the proportion of variance explained in principal component analysiscited by 3
  19. 2024Statistical Test for Feature Selection Pipelines by Selective Inferencecited by 3
  20. 2025Quantifying Statistical Significance of Deep Nearest Neighbor Anomaly Detection via Selective Inferencecited by 3
  21. 2019Selective inference after feature selection via multiscale bootstrapcited by 2
  22. 2022Transformer-based Personalized Attention Mechanism for Medical Images with Clinical Recordscited by 2
  23. 2024Rethinking Attention-Based Multiple Instance Learning for Whole-Slide Pathological Image Classification: An Instance Attribute Viewpointcited by 2
  24. 2022Selective Inference for Sparse Multitask Regression with Applications in Neuroimagingcited by 1
  25. 2025Statistical Test for Saliency Maps of Graph Neural Networks via Selective Inferencecited by 1

Built on by

Not cited yet.