ALICE: Learning a General-Purpose Pathology Foundation Model from Vision, Vision-Language, and Slide-Level Experts

2026-07-10

AI Summary

Existing pathology foundation models exhibit fragmented capabilities—spanning visual analysis, vision–language understanding, and whole-slide interpretation—due to disparities in pretraining objectives, data sources, and spatial scales, which necessitate separate specialized models and hinder unified deployment. To address this, this work proposes a multi-stage aggregation distillation framework that progressively transfers knowledge from eight expert models into distinct components of a single modular backbone network. This approach achieves, for the first time, a unified integration of multimodal and multiscale pathological knowledge within a general-purpose foundation model. Comprehensive evaluation across 21 task categories, 96 downstream tasks, and 48 datasets demonstrates that the resulting model attains state-of-the-art average performance in region-of-interest analysis, vision–language comprehension, and whole-slide clinical decision-making.

Abstract

Foundation models are reshaping computational pathology, yet their capabilities remain shaped by pretraining objectives, data sources, and spatial scales, fragmenting complementary expertise across separate backbones. Here we present ALICE, a unified foundation model trained through multi-stage agglomerative distillation that sequentially distills eight vision-only, vision-language, and slide-level teacher models into dedicated modules of a single backbone. ALICE is pretrained on 24,985,184 tile-level pathology images and 155,604 high-resolution images, and evaluated across 21 task scenarios, 96 downstream tasks, and 48 data sources, spanning region-of-interest tissue analysis, vision-language multimodal evaluation, and whole-slide clinical assessment. In all three evaluation settings, ALICE achieved the best average rank among task-matched pathology foundation models. These results demonstrate that agglomerative distillation can consolidate complementary capabilities from specialized models into a unified backbone for broad computational pathology applications. The model is available at https://github.com/WonderLandxD/ALICE.

Problem

• computational pathology
• foundation model
• multi-stage distillation
• vision-language
• whole-slide analysis

Innovation

• agglomerative distillation
• foundation model
• computational pathology
• vision-language learning
• multi-stage knowledge distillation