EXAONE Path 2.0: Pathology Foundation Model with End-to-End Supervision

Fuente: arXiv
Guardado en:
Detalles Bibliográficos
Autores principales: Pyeon, Myeongjang, Lee, Janghyeon, Lee, Minsoo, Yun, Juseung, Choi, Hwanil, Kim, Jonghyun, Kim, Jiwon, Hu, Yi, Jang, Jongseong, Lee, Soonyoung
Formato: Preprint
Publicado: 2025
Materias:
Acceso en línea:
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866911105073283072
author Pyeon, Myeongjang
Lee, Janghyeon
Lee, Minsoo
Yun, Juseung
Choi, Hwanil
Kim, Jonghyun
Kim, Jiwon
Hu, Yi
Jang, Jongseong
Lee, Soonyoung
author_facet Pyeon, Myeongjang
Lee, Janghyeon
Lee, Minsoo
Yun, Juseung
Choi, Hwanil
Kim, Jonghyun
Kim, Jiwon
Hu, Yi
Jang, Jongseong
Lee, Soonyoung
contents 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.
format Preprint
id arxiv_https___arxiv_org_abs_2507_06639
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle EXAONE Path 2.0: Pathology Foundation Model with End-to-End Supervision
Pyeon, Myeongjang
Lee, Janghyeon
Lee, Minsoo
Yun, Juseung
Choi, Hwanil
Kim, Jonghyun
Kim, Jiwon
Hu, Yi
Jang, Jongseong
Lee, Soonyoung
Computer Vision and Pattern Recognition
Artificial Intelligence
Machine Learning
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.
title EXAONE Path 2.0: Pathology Foundation Model with End-to-End Supervision
topic Computer Vision and Pattern Recognition
Artificial Intelligence
Machine Learning
url https://arxiv.org/abs/2507.06639