# Normal Guidance is what Attention Needs

arXiv: https://arxiv.org/abs/2605.27306 · HTML: https://mlsanity.com/paper/2605.27306/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 22 of its 38 references that are in the map.
- First posted on arXiv 26 May 2026.

## Builds on (most cited first, 22 of 22)

- [An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale](https://mlsanity.com/paper/2010.11929/) (2020)
- [Attention-based Deep Multiple Instance Learning](https://mlsanity.com/paper/1802.04712/) (2018)
- [ChestX-ray8: Hospital-scale Chest X-ray Database and Benchmarks on Weakly-Supervised Classification and Localization of Common Thorax Diseases](https://mlsanity.com/paper/1705.02315/) (2017)
- [Attention is not Explanation](https://mlsanity.com/paper/1902.10186/) (2019)
- [Data Efficient and Weakly Supervised Computational Pathology on Whole Slide Images](https://mlsanity.com/paper/2004.09666/) (2020)
- [Real-world Anomaly Detection in Surveillance Videos](https://mlsanity.com/paper/1801.04264/) (2018)
- [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)
- [Cluster-to-Conquer: A Framework for End-to-End Multi-Instance Learning for Whole Slide Image Classification](https://mlsanity.com/paper/2103.10626/) (2021)
- [Patched Diffusion Models for Unsupervised Anomaly Detection in Brain MRI](https://mlsanity.com/paper/2303.03758/) (2023)
- [Deep Multi-instance Networks with Sparse Label Assignment for Whole Mammogram Classification](https://mlsanity.com/paper/1612.05968/) (2016)
- [Modeling Concentrated Cross-Attention for Neural Machine Translation with Gaussian Mixture Model](https://mlsanity.com/paper/2109.05244/) (2021)
- [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)
- [Sm: enhanced localization in Multiple Instance Learning for medical imaging classification](https://mlsanity.com/paper/2410.03276/) (2024)
- [Detecting Heart Disease from Multi-View Ultrasound Images via Supervised Attention Multiple Instance Learning](https://mlsanity.com/paper/2306.00003/) (2023)
- [Multi-head Attention-based Deep Multiple Instance Learning](https://mlsanity.com/paper/2404.05362/) (2024)
- [Semantics-Aware Attention Guidance for Diagnosing Whole Slide Images](https://mlsanity.com/paper/2404.10894/) (2024)
- [PSA-MIL: A Probabilistic Spatial Attention-Based Multiple Instance Learning for Whole Slide Image Classification](https://mlsanity.com/paper/2503.16284/) (2025)
- [Patch2Loc: Learning to Localize Patches for Unsupervised Brain Lesion Detection](https://mlsanity.com/paper/2506.22504/) (2025)
- [Synthetic Data Reveals Generalization Gaps in Correlated Multiple Instance Learning](https://mlsanity.com/paper/2510.25759/) (2025)
- [A Multi-Dataset Benchmark of Multiple Instance Learning for 3D Neuroimage Classification](https://mlsanity.com/paper/2604.26807/) (2026)

## Abstract (arXiv)

We consider training classifiers for 3D medical images using only one binary label for the entire volume rather than a label for each 2D slice. In such weakly supervised settings, can we learn accurate classifiers for slice-level predictions? Attention-based multiple instance learning (MIL) can produce an attention score for every slice. Yet recent work demonstrates that a simple center-focused baseline that ignores image content can outperform attention-based and transformer-based MIL at slice-level classification of 3D brain scans. We show this baseline also outperforms existing MIL at slice-level classification of thoracic and abdominal CT scans. Motivated by this baseline, we propose Normal Guidance, a regularization technique that encourages the learned attention distribution to follow a bell-shaped curve. Across three medical imaging datasets totaling over 4 million 2D slices, we show our Normal Guidance enables attention-based and transformer-based MIL methods to deliver significantly better slice-level localization than the state-of-the-art while remaining competitive at whole-scan classification.