STELLA: Self-Evolving LLM Agent for Biomedical Research

Fuente: arXiv
Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Jin, Ruofan, Zhang, Zaixi, Wang, Mengdi, Cong, Le
Format: Preprint
Veröffentlicht: 2025
Schlagworte:
Online-Zugang:
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866908431387983872
author Jin, Ruofan
Zhang, Zaixi
Wang, Mengdi
Cong, Le
author_facet Jin, Ruofan
Zhang, Zaixi
Wang, Mengdi
Cong, Le
contents The rapid growth of biomedical data, tools, and literature has created a fragmented research landscape that outpaces human expertise. While AI agents offer a solution, they typically rely on static, manually curated toolsets, limiting their ability to adapt and scale. Here, we introduce STELLA, a self-evolving AI agent designed to overcome these limitations. STELLA employs a multi-agent architecture that autonomously improves its own capabilities through two core mechanisms: an evolving Template Library for reasoning strategies and a dynamic Tool Ocean that expands as a Tool Creation Agent automatically discovers and integrates new bioinformatics tools. This allows STELLA to learn from experience. We demonstrate that STELLA achieves state-of-the-art accuracy on a suite of biomedical benchmarks, scoring approximately 26\% on Humanity's Last Exam: Biomedicine, 54\% on LAB-Bench: DBQA, and 63\% on LAB-Bench: LitQA, outperforming leading models by up to 6 percentage points. More importantly, we show that its performance systematically improves with experience; for instance, its accuracy on the Humanity's Last Exam benchmark almost doubles with increased trials. STELLA represents a significant advance towards AI Agent systems that can learn and grow, dynamically scaling their expertise to accelerate the pace of biomedical discovery.
format Preprint
id arxiv_https___arxiv_org_abs_2507_02004
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle STELLA: Self-Evolving LLM Agent for Biomedical Research
Jin, Ruofan
Zhang, Zaixi
Wang, Mengdi
Cong, Le
Artificial Intelligence
Computation and Language
Biomolecules
The rapid growth of biomedical data, tools, and literature has created a fragmented research landscape that outpaces human expertise. While AI agents offer a solution, they typically rely on static, manually curated toolsets, limiting their ability to adapt and scale. Here, we introduce STELLA, a self-evolving AI agent designed to overcome these limitations. STELLA employs a multi-agent architecture that autonomously improves its own capabilities through two core mechanisms: an evolving Template Library for reasoning strategies and a dynamic Tool Ocean that expands as a Tool Creation Agent automatically discovers and integrates new bioinformatics tools. This allows STELLA to learn from experience. We demonstrate that STELLA achieves state-of-the-art accuracy on a suite of biomedical benchmarks, scoring approximately 26\% on Humanity's Last Exam: Biomedicine, 54\% on LAB-Bench: DBQA, and 63\% on LAB-Bench: LitQA, outperforming leading models by up to 6 percentage points. More importantly, we show that its performance systematically improves with experience; for instance, its accuracy on the Humanity's Last Exam benchmark almost doubles with increased trials. STELLA represents a significant advance towards AI Agent systems that can learn and grow, dynamically scaling their expertise to accelerate the pace of biomedical discovery.
title STELLA: Self-Evolving LLM Agent for Biomedical Research
topic Artificial Intelligence
Computation and Language
Biomolecules
url https://arxiv.org/abs/2507.02004