EXAONE Path 2.0: Pathology Foundation Model with End-to-End Supervision
This work addresses data inefficiency and domain-specific feature limitations in pathology AI for biomarker prediction, representing a novel method rather than an incremental improvement.
The paper tackles the challenge of handling gigapixel whole-slide images in digital pathology by introducing EXAONE Path 2.0, a foundation model that uses end-to-end slide-level supervision to learn patch-level representations, achieving state-of-the-art average performance across 10 biomarker prediction tasks with only 37k WSIs for training.
In digital pathology, whole-slide images (WSIs) are often difficult to handle due to their gigapixel scale, so most approaches train patch encoders via self-supervised learning (SSL) and then aggregate the patch-level embeddings via multiple instance learning (MIL) or slide encoders for downstream tasks. However, patch-level SSL may overlook complex domain-specific features that are essential for biomarker prediction, such as mutation status and molecular characteristics, as SSL methods rely only on basic augmentations selected for natural image domains on small patch-level area. Moreover, SSL methods remain less data efficient than fully supervised approaches, requiring extensive computational resources and datasets to achieve competitive performance. To address these limitations, we present EXAONE Path 2.0, a pathology foundation model that learns patch-level representations under direct slide-level supervision. Using only 37k WSIs for training, EXAONE Path 2.0 achieves state-of-the-art average performance across 10 biomarker prediction tasks, demonstrating remarkable data efficiency.