# Cross-Modal Prototype Allocation: Unsupervised Slide Representation Learning via Patch-Text Contrast in Computational Pathology

arXiv: https://arxiv.org/abs/2503.20190 · HTML: https://mlsanity.com/paper/2503.20190/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 17 of its 26 references that are in the map.
- First posted on arXiv 26 Mar 2025.

## Builds on (most cited first, 17 of 17)

- [Attention-based Deep Multiple Instance Learning](https://mlsanity.com/paper/1802.04712/) (2018)
- [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)
- [Visual Language Pretrained Multiple Instance Zero-Shot Transfer for Histopathology Images](https://mlsanity.com/paper/2306.07831/) (2023)
- [Whole Slide Images are 2D Point Clouds: Context-Aware Survival Prediction using Patch-based Graph Convolutional Networks](https://mlsanity.com/paper/2107.13048/) (2021)
- [Morphological Prototyping for Unsupervised Slide Representation Learning in Computational Pathology](https://mlsanity.com/paper/2405.11643/) (2024)
- [Feature Re-Embedding: Towards Foundation Model-Level Performance in Computational Pathology](https://mlsanity.com/paper/2402.17228/) (2024)
- [Dynamic Graph Representation with Knowledge-aware Attention for Histopathology Whole Slide Image Analysis](https://mlsanity.com/paper/2403.07719/) (2024)
- [Handcrafted Histological Transformer (H2T): Unsupervised Representation of Whole Slide Images](https://mlsanity.com/paper/2202.07001/) (2022)
- [Shapley Values-enabled Progressive Pseudo Bag Augmentation for Whole Slide Image Classification](https://mlsanity.com/paper/2312.05490/) (2023)
- [Agent Aggregator with Mask Denoise Mechanism for Histopathology Whole Slide Image Analysis](https://mlsanity.com/paper/2409.11664/) (2024)
- [RetMIL: Retentive Multiple Instance Learning for Histopathological Whole Slide Image Classification](https://mlsanity.com/paper/2403.10858/) (2024)
- [MergeUp-augmented Semi-Weakly Supervised Learning for WSI Classification](https://mlsanity.com/paper/2408.12825/) (2024)
- [Feature Re-calibration based Multiple Instance Learning for Whole Slide Image Classification](https://mlsanity.com/paper/2206.10878/) (2022)
- [Towards a Comprehensive Benchmark for Pathological Lymph Node Metastasis in Breast Cancer Sections](https://mlsanity.com/paper/2411.10752/) (2024)
- [Dynamic Hypergraph Representation for Bone Metastasis Cancer Analysis](https://mlsanity.com/paper/2501.16787/) (2025)

## Abstract (arXiv)

With the rapid advancement of pathology foundation models (FMs), the representation learning of whole slide images (WSIs) attracts increasing attention. Existing studies develop high-quality patch feature extractors and employ carefully designed aggregation schemes to derive slide-level representations. However, mainstream weakly supervised slide representation learning methods, primarily based on multiple instance learning (MIL), are tailored to specific downstream tasks, which limits their generalizability. To address this issue, some studies explore unsupervised slide representation learning. However, these approaches focus solely on the visual modality of patches, neglecting the rich semantic information embedded in textual data. In this work, we propose ProAlign, a cross-modal unsupervised slide representation learning framework. Specifically, we leverage a large language model (LLM) to generate descriptive text for the prototype types present in a WSI, introducing patch-text contrast to construct initial prototype embeddings. Furthermore, we propose a parameter-free attention aggregation strategy that utilizes the similarity between patches and these prototypes to form unsupervised slide embeddings applicable to a wide range of downstream tasks. Extensive experiments on four public datasets show that ProAlign outperforms existing unsupervised frameworks and achieves performance comparable to some weakly supervised models.