Newest papers

BatchBench: Benchmarking AI Robustness to Batch Effects and Domain Shift in Biomedical Images - A JUMP Cell Painting Leaderboard

Sonja Topf, Ana Sanchez-Fernandez, Gnter Klambauer · 2026 · arXiv (Cornell University)

DOI: 10.48550/arxiv.2610.30104 · cited by 0 works (OpenAlex) · open-access PDF

Indonesian

Meskipun kemajuan baru-baru ini, sistem AI saat ini tetap sangat rentan terhadap perubahan domain, terutama dalam pencitraan biomedis, di mana gambar biasanya diperoleh dalam kondisi eksperimental yang berbeda secara substansial. Meskipun metode penyesuaian domain baru-baru ini telah membuat kemajuan yang signifikan, kesenjangan kinerja yang signifikan selama perubahan domain tetap ada. Untuk mengukur kemajuan dalam hal ini, kami memperkenalkan BatchBench, sebuah panel terkemuka yang mengevaluasi dan membandingkan ketahanan sistem AI pada gambar biomedis dari batch eksperimental dan sumber data yang tidak terlihat. BatchBench dibangun pada JUMP-CP, salah satu dataset pencitraan biomedis yang terbesar yang tersedia secara publik.

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

Abstract

Despite recent advances, current AI systems remain highly susceptible to domain shifts, particularly in biomedical imaging, where images are usually acquired under substantially different experimental conditions. Although recent domain adaptation methods have made substantial progress, significant performance gaps under domain shift persist. To measure progress in this regard, we introduce BatchBench, a leaderboard that evaluates and compares the robustness of AI systems on biomedical images from unseen experimental batches and data sources. BatchBench is built on JUMP-CP, one of the largest publicly available biomedical imaging datasets. Generated using the Cell Painting assay, JUMP-CP exhibits pronounced batch effects and domain shifts, making it a challenging testbed for evaluating robustness to distributional changes.

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

Topics: Robustness (evolution), Computer science, Benchmarking, Artificial intelligence, Testbed, Domain adaptation

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