# From Correlation to Causation: Max-Pooling-Based Multi-Instance Learning Leads to More Robust Whole Slide Image Classification

arXiv: https://arxiv.org/abs/2408.09449 · HTML: https://mlsanity.com/paper/2408.09449/

## Key facts

- Cited by 7 papers in the map (data through 14 Oct 2026).
- Builds on 27 of its 55 references that are in the map.
- First posted on arXiv 18 Aug 2024.

## Builds on (most cited first, 27 of 27)

- [Attention Is All You Need](https://mlsanity.com/paper/1706.03762/) (2017)
- [Deep Residual Learning for Image Recognition](https://mlsanity.com/paper/1512.03385/) (2015)
- [Attention-based Deep Multiple Instance Learning](https://mlsanity.com/paper/1802.04712/) (2018)
- [Deep Variational Information Bottleneck](https://mlsanity.com/paper/1612.00410/) (2016)
- [Deep Learning and the Information Bottleneck Principle](https://mlsanity.com/paper/1503.02406/) (2015)
- [Data Efficient and Weakly Supervised Computational Pathology on Whole Slide Images](https://mlsanity.com/paper/2004.09666/) (2020)
- [TransMIL: Transformer based Correlated Multiple Instance Learning for Whole Slide Image Classification](https://mlsanity.com/paper/2106.00908/) (2021)
- [Dual-stream Multiple Instance Learning Network for Whole Slide Image Classification with Self-supervised Contrastive Learning](https://mlsanity.com/paper/2011.08939/) (2020)
- [Self-Supervised Learning with Data Augmentations Provably Isolates Content from Style](https://mlsanity.com/paper/2106.04619/) (2021)
- [DTFD-MIL: Double-Tier Feature Distillation Multiple Instance Learning for Histopathology Whole Slide Image Classification](https://mlsanity.com/paper/2203.12081/) (2022)
- [Whole Slide Images based Cancer Survival Prediction using Attention Guided Deep Multiple Instance Learning Networks](https://mlsanity.com/paper/2009.11169/) (2020)
- [Whole Slide Images are 2D Point Clouds: Context-Aware Survival Prediction using Patch-based Graph Convolutional Networks](https://mlsanity.com/paper/2107.13048/) (2021)
- [Revisiting Multiple Instance Neural Networks](https://mlsanity.com/paper/1610.02501/) (2016)
- [Interventional Bag Multi-Instance Learning On Whole-Slide Pathological Images](https://mlsanity.com/paper/2303.06873/) (2023)
- [Feature Re-Embedding: Towards Foundation Model-Level Performance in Computational Pathology](https://mlsanity.com/paper/2402.17228/) (2024)
- [Attention-Challenging Multiple Instance Learning for Whole Slide Image Classification](https://mlsanity.com/paper/2311.07125/) (2023)
- [Multiple Instance Learning Framework with Masked Hard Instance Mining for Whole Slide Image Classification](https://mlsanity.com/paper/2307.15254/) (2023)
- [Additive MIL: Intrinsically Interpretable Multiple Instance Learning for Pathology](https://mlsanity.com/paper/2206.01794/) (2022)
- [Task-specific Fine-tuning via Variational Information Bottleneck for Weakly-supervised Pathology Whole Slide Image Classification](https://mlsanity.com/paper/2303.08446/) (2023)
- [Bi-directional Weakly Supervised Knowledge Distillation for Whole Slide Image Classification](https://mlsanity.com/paper/2210.03664/) (2022)
- [Benchmarking the Robustness of Deep Neural Networks to Common Corruptions in Digital Pathology](https://mlsanity.com/paper/2206.14973/) (2022)
- [Inherently Interpretable Time Series Classification via Multiple Instance Learning](https://mlsanity.com/paper/2311.10049/) (2023)
- [Rethinking Multiple Instance Learning for Whole Slide Image Classification: A Good Instance Classifier is All You Need](https://mlsanity.com/paper/2307.02249/) (2023)
- [Multi-Instance Causal Representation Learning for Instance Label Prediction and Out-of-Distribution Generalization](https://mlsanity.com/paper/2202.12570/) (2022)
- [Memorization in deep learning: A survey](https://mlsanity.com/paper/2406.03880/) (2024)
- [Reproducibility in Multiple Instance Learning: A Case For Algorithmic Unit Tests](https://mlsanity.com/paper/2310.17867/) (2023)
- [AEM: Attention Entropy Maximization for Multiple Instance Learning based Whole Slide Image Classification](https://mlsanity.com/paper/2406.15303/) (2024)

## Built on by (most cited first, 7 of 7)

- [VistaWise: Building Cost-Effective Agent with Cross-Modal Knowledge Graph for Minecraft](https://mlsanity.com/paper/2508.18722/) (2025)
- [VideoStir: Understanding Long Videos via Spatio-Temporally Structured and Intent-Aware RAG](https://mlsanity.com/paper/2604.05418/) (2026)
- [TRACE: Trajectory Recovery for Continuous Mechanism Evolution in Causal Representation Learning](https://mlsanity.com/paper/2601.21135/) (2026)
- [Multi-View Stenosis Classification Leveraging Transformer-Based Multiple-Instance Learning Using Real-World Clinical Data](https://mlsanity.com/paper/2602.02067/) (2026)
- [The Weakest Link Tells It All: Outcome-Supervised Process Reward Modeling via Learnable Credit Assignment](https://mlsanity.com/paper/2606.27739/) (2026)
- [From Patches to Evidence Balls: Class-Conditioned Evidence Retrieval for Few-Shot Whole Slide Image Classification](https://mlsanity.com/paper/2608.01104/) (2026)
- [Weakly-supervised Kidney Tumor Classification from CT Scans with Multi-Instance Learning and Anatomical Filtering](https://mlsanity.com/paper/2609.07178/) (2026)

## Abstract (arXiv)

In whole slide images (WSIs) analysis, attention-based multi-instance learning (MIL) models are susceptible to spurious correlations and degrade under domain shift. These methods may assign high attention weights to non-tumor regions, such as staining biases or artifacts, leading to unreliable tumor region localization. In this paper, we revisit max-pooling-based MIL methods from a causal perspective. Under mild assumptions, our theoretical results demonstrate that max-pooling encourages the model to focus on causal factors while ignoring bias-related factors. Furthermore, we discover that existing max-pooling-based methods may overfit the training set through rote memorization of instance features and fail to learn meaningful patterns. To address these issues, we propose FocusMIL, which couples max-pooling with an instance-level variational information bottleneck (VIB) to learn compact, predictive latent representations, and employs a multi-bag mini-batch scheme to stabilize optimization. We conduct comprehensive experiments on three real-world datasets and one semi-synthetic dataset. The results show that, by capturing causal factors, FocusMIL exhibits significant advantages in out-of-distribution scenarios and instance-level tumor region localization tasks.