mlsanityData to 14 Oct 2026

Synthetic Data Reveals Generalization Gaps in Correlated Multiple Instance Learning

arXiv 2510.25759 ↗First posted 29 Oct 2025cs.LGcited by 1

Not on the map. Why?

Abstract (arXiv)

Multiple instance learning (MIL) is often used in medical imaging to classify high-resolution 2D images by processing patches or classify 3D volumes by processing slices. However, conventional MIL approaches treat instances separately, ignoring contextual relationships such as the appearance of nearby patches or slices that can be essential in real applications. We design a synthetic classification task where accounting for adjacent instance features is crucial for accurate prediction. We demonstrate the limitations of off-the-shelf MIL approaches by quantifying their performance compared to the optimal Bayes estimator for this task, which is available in closed-form. We empirically show that newer correlated MIL methods still do not achieve the best possible performance when trained with ten thousand training samples, each containing many instances.

Key facts. Cited by 1 paper in the map (data through 14 Oct 2026). Builds on 13 of its 23 references that are in the map. First posted on arXiv 29 Oct 2025.

Builds on

13 of 23 references.

References · 13

  1. 2020An Image is Worth 16x16 Words: Transformers for Image Recognition at Scalecited by 30,186
  2. 2021Swin Transformer: Hierarchical Vision Transformer using Shifted Windowscited by 12,139
  3. 2017Quo Vadis, Action Recognition? A New Model and the Kinetics Datasetcited by 3,270
  4. 2018Attention-based Deep Multiple Instance Learningcited by 1,095
  5. 2020Data Efficient and Weakly Supervised Computational Pathology on Whole Slide Imagescited by 745
  6. 2020Dual-stream Multiple Instance Learning Network for Whole Slide Image Classification with Self-supervised Contrastive Learningcited by 478
  7. 2016Deep Multi-instance Networks with Sparse Label Assignment for Whole Mammogram Classificationcited by 32
  8. 2025Do Multiple Instance Learning Models Transfer?cited by 29
  9. 2023Adapting Pre-trained Vision Transformers from 2D to 3D through Weight Inflation Improves Medical Image Segmentationcited by 11
  10. 2023Reproducibility in Multiple Instance Learning: A Case For Algorithmic Unit Testscited by 11
  11. 2024Sm: enhanced localization in Multiple Instance Learning for medical imaging classificationcited by 10
  12. 2024Multi-head Attention-based Deep Multiple Instance Learningcited by 6
1 more
  1. 2023A Probabilistic Method to Predict Classifier Accuracy on Larger Datasets given Small Pilot Datacited by 3

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2026: 1.

1 paper.

  1. 2026Normal Guidance is what Attention Needscited by 0