# In-Context Multiple Instance Learning

arXiv: https://arxiv.org/abs/2606.06458 · HTML: https://mlsanity.com/paper/2606.06458/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 28 of its 50 references that are in the map.
- First posted on arXiv 4 Jun 2026.

## Builds on (most cited first, 28 of 28)

- [PyTorch: An Imperative Style, High-Performance Deep Learning Library](https://mlsanity.com/paper/1912.01703/) (2019)
- [Deep Sets](https://mlsanity.com/paper/1703.06114/) (2017)
- [Perceiver: General Perception with Iterative Attention](https://mlsanity.com/paper/2103.03206/) (2021)
- [Attention-based Deep Multiple Instance Learning](https://mlsanity.com/paper/1802.04712/) (2018)
- [TabPFN: A Transformer That Solves Small Tabular Classification Problems in a Second](https://mlsanity.com/paper/2207.01848/) (2022)
- [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)
- [Transformers Can Do Bayesian Inference](https://mlsanity.com/paper/2112.10510/) (2021)
- [TabICL: A Tabular Foundation Model for In-Context Learning on Large Data](https://mlsanity.com/paper/2502.05564/) (2025)
- [Multiple Instance Learning: A Survey of Problem Characteristics and Applications](https://mlsanity.com/paper/1612.03365/) (2016)
- [The Road Less Scheduled](https://mlsanity.com/paper/2405.15682/) (2024)
- [Multimodal Whole Slide Foundation Model for Pathology](https://mlsanity.com/paper/2411.19666/) (2024)
- [Visual Language Pretrained Multiple Instance Zero-Shot Transfer for Histopathology Images](https://mlsanity.com/paper/2306.07831/) (2023)
- [PFNs4BO: In-Context Learning for Bayesian Optimization](https://mlsanity.com/paper/2305.17535/) (2023)
- [Attention-Challenging Multiple Instance Learning for Whole Slide Image Classification](https://mlsanity.com/paper/2311.07125/) (2023)
- [From Tables to Time: Extending TabPFN-v2 to Time Series Forecasting](https://mlsanity.com/paper/2501.02945/) (2025)
- [Do-PFN: In-Context Learning for Causal Effect Estimation](https://mlsanity.com/paper/2506.06039/) (2025)
- [Real-TabPFN: Improving Tabular Foundation Models via Continued Pre-training With Real-World Data](https://mlsanity.com/paper/2507.03971/) (2025)
- [Parameterized Machine Learning for High-Energy Physics](https://mlsanity.com/paper/1601.07913/) (2016)
- [Do Multiple Instance Learning Models Transfer?](https://mlsanity.com/paper/2506.09022/) (2025)
- [xMIL: Insightful Explanations for Multiple Instance Learning in Histopathology](https://mlsanity.com/paper/2406.04280/) (2024)
- [TabPFN-Wide: Continued Pre-Training for Extreme Feature Counts](https://mlsanity.com/paper/2510.06162/) (2025)
- [nanoTabPFN: A Lightweight and Educational Reimplementation of TabPFN](https://mlsanity.com/paper/2511.03634/) (2025)
- [Tools and Practices for Responsible AI Engineering](https://mlsanity.com/paper/2201.05647/) (2022)
- [MeDi: Metadata-Guided Diffusion Models for Mitigating Biases in Tumor Classification](https://mlsanity.com/paper/2506.17140/) (2025)
- [MapPFN: Learning Causal Perturbation Maps in Context](https://mlsanity.com/paper/2601.21092/) (2026)
- [torchmil: A PyTorch-based library for deep Multiple Instance Learning](https://mlsanity.com/paper/2509.08129/) (2025)
- [Beyond Attention Heatmaps: How to Get Better Explanations for Multiple Instance Learning Models in Histopathology](https://mlsanity.com/paper/2603.08328/) (2026)

## Abstract (arXiv)

Multiple Instance Learning (MIL) addresses problems where supervision is available at the level of bags of instances and has been successfully applied in fields ranging from computational pathology to satellite imagery. Nevertheless, existing algorithms struggle in the low-label regime that characterizes many real-world applications. Flexible models overfit and rigid ones fail to adapt to the task at hand. We show that pretraining an in-context learner with a Perceiver-style architecture on synthetic data yields a model that can solve new tasks from a handful of labeled bags. At inference time, classification happens in a single forward pass and requires no gradient updates. We propose and investigate different synthetic data generators for bag-structured data and find that they capture complementary inductive biases. A model pretrained on a mixture of these generators inherits their per-task strengths and achieves the best average performance across twelve MIL benchmarks, outperforming supervised baselines that require task-specific training.