# FFM-CP: Cross-Backbone Fusion of Vision-Language Foundation Models for Few-Shot Computational Pathology

arXiv: https://arxiv.org/abs/2609.27710 · HTML: https://mlsanity.com/paper/2609.27710/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Placed in Medical vision-language models, one of 16 directions in Multimodal LLMs / vision-language models.
- Builds on 16 of its 39 references that are in the map.
- First posted on arXiv 23 Sep 2026.

## Placed in

- Multimodal LLMs / vision-language models / [Medical vision-language models](https://mlsanity.com/mllm/medical-vision-language-models/)

## Builds on (most cited first, 16 of 16)

- [Learning Transferable Visual Models From Natural Language Supervision](https://mlsanity.com/paper/2103.00020/) (2021)
- [Conditional Prompt Learning for Vision-Language Models](https://mlsanity.com/paper/2203.05557/) (2022)
- [CLIP-Adapter: Better Vision-Language Models with Feature Adapters](https://mlsanity.com/paper/2110.04544/) (2021)
- [Word Translation Without Parallel Data](https://mlsanity.com/paper/1710.04087/) (2017)
- [Tip-Adapter: Training-free CLIP-Adapter for Better Vision-Language Modeling](https://mlsanity.com/paper/2111.03930/) (2021)
- [Quilt-1M: One Million Image-Text Pairs for Histopathology](https://mlsanity.com/paper/2306.11207/) (2023)
- [BRACS: A Dataset for BReAst Carcinoma Subtyping in H&E Histology Images](https://mlsanity.com/paper/2111.04740/) (2021)
- [BACH: Grand Challenge on Breast Cancer Histology Images](https://mlsanity.com/paper/1808.04277/) (2018)
- [GraphAdapter: Tuning Vision-Language Models With Dual Knowledge Graph](https://mlsanity.com/paper/2309.13625/) (2023)
- [ViLa-MIL: Dual-scale Vision-Language Multiple Instance Learning for Whole Slide Image Classification](https://mlsanity.com/paper/2502.08391/) (2025)
- [Knowledge-enhanced Pretraining for Vision-language Pathology Foundation Model on Cancer Diagnosis](https://mlsanity.com/paper/2412.13126/) (2024)
- [FOCUS: Knowledge-enhanced Adaptive Visual Compression for Few-shot Whole Slide Image Classification](https://mlsanity.com/paper/2411.14743/) (2024)
- [MSCPT: Few-shot Whole Slide Image Classification with Multi-scale and Context-focused Prompt Tuning](https://mlsanity.com/paper/2408.11505/) (2024)
- [Exploring Cross-Modal Flows for Few-Shot Learning](https://mlsanity.com/paper/2510.14543/) (2025)
- [Fusion of Multi-scale Heterogeneous Pathology Foundation Models for Whole Slide Image Analysis](https://mlsanity.com/paper/2510.27237/) (2025)
- [Plug-and-Play Logit Fusion for Heterogeneous Pathology Foundation Models](https://mlsanity.com/paper/2604.07779/) (2026)

## Abstract (arXiv)

Pathology vision-language foundation models vary in performance across diseases and tasks, with no single model consistently performing best. The high cost of expert pathology annotation can also limit the labeled data available for task-specific adaptation. Combining complementary pretrained representations is a potential approach to these limitations, yet learning an effective fusion from few labeled examples remains challenging. We introduce Few-shot Fusion Foundation Models of Computational Pathology (FFM-CP), which is a framework that combines multiple pathology vision-language models in the few-shot learning setting. The framework first aligns heterogeneous representations using a closed-form Orthogonal Procrustes transformation estimated from corresponding support images. This alignment preserves within-model feature geometry without training an additional alignment network. Within the aligned space, a unified graph enables information exchange across backbones by jointly refining support-image features and visual and textual class prototypes. These refined representations support complementary text-prototype and case-retrieval branches that capture semantic class knowledge and within-class visual variation, respectively. Each branch learns to combine predictions from all ordered backbone pairs, allowing queries encoded by one model to draw on evidence represented by another. We evaluate three backbone combinations on six histopathology datasets at 4, 8, and 16 shots per class. FFM-CP achieves higher mean macro-F1 than the strongest individually adapted member of each fused set in 50 of 54 comparisons. These findings suggest that combining complementary pretrained representations can improve histopathological classification when annotations are limited.