mlsanityData to 14 Oct 2026

GCE-MIL: Faithful and Recoverable Evidence for Multiple Instance Learning in Whole-Slide Imaging

arXiv 2605.17456 ↗First posted 17 May 2026cs.CV cs.AIcited by 0

Not on the map. Why?

Abstract (arXiv)

Multiple instance learning (MIL) is the standard approach for whole-slide image (WSI) classification and survival prediction, where attention-based models ag gregate patch features into slide-level predictions. These models treat attention weights as evidence for their predictions, but attention is optimized for classi fication, not for identifying which patches actually support the diagnosis. This conflation leads to three failures: selected patches are insufficient (keeping them alone drops Macro-F1 by 0.078), unnecessary (removing them barely changes the prediction), and unrecoverable (continuous attention scores disagree with discrete patch subsets used at inference). The central premise is that evidence quality should be optimized directly through explicit criteria- Sufficiency, Necessity, and Recov erability (S/N/R)- rather than inherited as a byproduct of classification. GCE-MIL is a backbone-agnostic wrapper implemented through three injection modes and three evidence components: a grounding mechanism that aligns selection with domain-specific concepts, noisy-OR coverage that acts as a differentiable proxy for interventional evidence search, and threshold-plus-repair recovery that converts continuous selectors into discrete subsets through marginal-guided repair. Across 9 backbones and 9 datasets (81 configurations), GCE-MIL improves average Macro-F1 by 0.024 and C-index by 0.014, reduces the continuous-discrete gap by 4-7, and increases complement degradation by 2-4. With optional tile prefiltering after discrete recovery, inference runs up to 5 faster while retaining 0.989 full-bag utility.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Builds on 48 of its 79 references that are in the map. First posted on arXiv 17 May 2026.

Builds on

48 of 79 references.

References · 48

  1. 2015Deep Residual Learning for Image Recognitioncited by 52,096
  2. 2020A Simple Framework for Contrastive Learning of Visual Representationscited by 12,049
  3. 2015Distilling the Knowledge in a Neural Networkcited by 10,324
  4. 2014ImageNet Large Scale Visual Recognition Challengecited by 8,498
  5. 2023DINOv2: Learning Robust Visual Features without Supervisioncited by 7,974
  6. 2019Momentum Contrast for Unsupervised Visual Representation Learningcited by 7,513
  7. 2021Emerging Properties in Self-Supervised Vision Transformerscited by 6,968
  8. 2017Axiomatic Attribution for Deep Networkscited by 3,384
  9. 2016Categorical Reparameterization with Gumbel-Softmaxcited by 2,739
  10. 2013Deep Inside Convolutional Networks: Visualising Image Classification Models and Saliency Mapscited by 2,371
  11. 2013Visualizing and Understanding Convolutional Networkscited by 2,290
  12. 2017Interpretability Beyond Feature Attribution: Quantitative Testing with Concept Activation Vectors (TCAV)cited by 1,319
36 more
  1. 2016The Concrete Distribution: A Continuous Relaxation of Discrete Random Variablescited by 1,148
  2. 2018Attention-based Deep Multiple Instance Learningcited by 1,095
  3. 2020Concept Bottleneck Modelscited by 1,032
  4. 2018RISE: Randomized Input Sampling for Explanation of Black-box Modelscited by 838
  5. 2020Data Efficient and Weakly Supervised Computational Pathology on Whole Slide Imagescited by 745
  6. 2018This Looks Like That: Deep Learning for Interpretable Image Recognitioncited by 737
  7. 2017Interpretable Explanations of Black Boxes by Meaningful Perturbationcited by 612
  8. 2017Learning Sparse Neural Networks through L0 Regularizationcited by 583
  9. 2021TransMIL: Transformer based Correlated Multiple Instance Learning for Whole Slide Image Classificationcited by 555
  10. 2019ERASER: A Benchmark to Evaluate Rationalized NLP Modelscited by 530
  11. 2018Towards Robust Interpretability with Self-Explaining Neural Networkscited by 505
  12. 2020Dual-stream Multiple Instance Learning Network for Whole Slide Image Classification with Self-supervised Contrastive Learningcited by 478
  13. 2021Nystr"omformer: A Nystr"om-Based Algorithm for Approximating Self-Attentioncited by 471
  14. 2016From Softmax to Sparsemax: A Sparse Model of Attention and Multi-Label Classificationcited by 341
  15. 2022DTFD-MIL: Double-Tier Feature Distillation Multiple Instance Learning for Histopathology Whole Slide Image Classificationcited by 256
  16. 2018Learning to Explain: An Information-Theoretic Perspective on Model Interpretationcited by 248
  17. 2020Whole Slide Images based Cancer Survival Prediction using Attention Guided Deep Multiple Instance Learning Networkscited by 178
  18. 2024Multimodal Whole Slide Foundation Model for Pathologycited by 162
  19. 2022Benchmarking Self-Supervised Learning on Diverse Pathology Datasetscited by 150
  20. 2023Visual Language Pretrained Multiple Instance Zero-Shot Transfer for Histopathology Imagescited by 134
  21. 2021BRACS: A Dataset for BReAst Carcinoma Subtyping in H&E Histology Imagescited by 132
  22. 2021Whole Slide Images are 2D Point Clouds: Context-Aware Survival Prediction using Patch-based Graph Convolutional Networkscited by 124
  23. 2023Artificial Intelligence for Digital and Computational Pathologycited by 108
  24. 2023Interventional Bag Multi-Instance Learning On Whole-Slide Pathological Imagescited by 91
  25. 2024MambaMIL: Enhancing Long Sequence Modeling with Sequence Reordering in Computational Pathologycited by 88
  26. 2023Attention-Challenging Multiple Instance Learning for Whole Slide Image Classificationcited by 61
  27. 2023Multiple Instance Learning Framework with Masked Hard Instance Mining for Whole Slide Image Classificationcited by 60
  28. 2015Classifying and Segmenting Microscopy Images Using Convolutional Multiple Instance Learningcited by 38
  29. 2025Do Multiple Instance Learning Models Transfer?cited by 29
  30. 2023Rethinking Multiple Instance Learning for Whole Slide Image Classification: A Good Instance Classifier is All You Needcited by 22
  31. 2023HIGT: Hierarchical Interaction Graph-Transformer for Whole Slide Image Analysiscited by 17
  32. 2024AEM: Attention Entropy Maximization for Multiple Instance Learning based Whole Slide Image Classificationcited by 10
  33. 2025GECKO: Gigapixel Vision-Concept Contrastive Pretraining in Histopathologycited by 8
  34. 2019An attention-based multi-resolution model for prostate whole slide imageclassification and localizationcited by 7
  35. 2024PATHS: A Hierarchical Transformer for Efficient Whole Slide Image Analysiscited by 7
  36. 2026ASMIL: Attention-Stabilized Multiple Instance Learning for Whole Slide Imagingcited by 6

Built on by

Not cited yet.