An AI system to help scientists write expert-level empirical software
A new AI system called Empirical Research Assistance (ERA) has been developed to aid scientists in creating expert-level empirical software. Utilizing a Large Language Model and Tree Search, ERA enhances the quality of scientific software and has shown promising results across various fields. Notably, it outperformed human-developed methods in bioinformatics and epidemiology, indicating its potential to accelerate scientific discovery.
- ▪ERA uses advanced AI techniques to systematically improve the quality of scientific software.
- ▪The system has successfully discovered novel methods in bioinformatics that surpassed top human-developed methods.
- ▪ERA generated models for forecasting COVID-19 hospitalizations that outperformed existing models from the CDC.
Opening excerpt (first ~120 words) tap to expand
Computer Science > Artificial Intelligence arXiv:2509.06503 (cs) COVID-19 e-print Important: e-prints posted on arXiv are not peer-reviewed by arXiv; they should not be relied upon without context to guide clinical practice or health-related behavior and should not be reported in news media as established information without consulting multiple experts in the field. [Submitted on 8 Sep 2025 (v1), last revised 21 May 2026 (this version, v3)] Title:An AI system to help scientists write expert-level empirical software Authors:Eser Aygün, Anastasiya Belyaeva, Gheorghe Comanici, Marc Coram, Hao Cui, Jake Garrison, Renee Johnston Anton Kast, Cory Y. McLean, Peter Norgaard, Zahra Shamsi, David Smalling, James Thompson, Subhashini Venugopalan, Brian P.
…
Excerpt limited to ~120 words for fair-use compliance. The full article is at arXiv.org.
Discussion
0 commentsMore from arXiv.org
