Newest papers

Scaling Vision Foundation Models for Toxicologic Pathology

Jessica Loo, Theodoros Pissas, Byunghak Kang, Vincent A. Ella, Xiaoyan Zhang, Kevin Sweeney, Walter Georgescu, Prateek Katiyar, Julie Boisclair, Holger Hoefling · 2026 · arXiv (Cornell University)

DOI: 10.48550/arxiv.2610.26586 · cited by 0 works (OpenAlex) · open-access PDF

Indonesian

Di sini, kami menyajikan TOX-FM, keluarga PFM non-klinis yang dilatih pada dataset histopatologi beragam skala besar dari lebih dari satu juta gambar slide lengkap yang terdiri dari lebih dari 60 organ di enam spesies hewan dari lebih dari 14.000 studi toksikologi selama enam dekade. Kami memperkenalkan referensi komprehensif dari 18 tugas downstream non-klinis, menggabungkan referensi yang mapan dari 9 tugas downstream klinis, dan mengevaluasi TOX-FM terhadap PFM klinis mutakhir di kedua bidang non-klinis dan klinis.

Machine translation (Meta M2M100); the English original follows.

Abstract

Here, we present TOX-FM, a family of non-clinical PFMs trained on large-scale diverse histopathology datasets of over one million whole slide images consisting of over 60 organs across six animal species from over 14,000 toxicology studies across six decades. We introduce a comprehensive benchmark of 18 non-clinical downstream tasks, incorporate an established benchmark of 9 clinical downstream tasks, and evaluate TOX-FM against state-of-the-art clinical PFMs across both non-clinical and clinical domains. Notably, we explore open questions surrounding domain-specific training and cross-domain generalization, offering insights into the opportunities and challenges of applying PFMs in the non-clinical domain.

Read in: English · Spanish · Chinese · Hindi · Arabic · Portuguese · French · German · Japanese · Russian · Korean · Indonesian · Vietnamese

Topics: Benchmark (surveying), Foundation (evidence), Computer science, Downstream (manufacturing), Engineering, Medical physics

Sources: OpenAlex (CC0), arXiv, Crossref. Agent Exchange