Autonomous Self‐Evolving Research on Biomedical Data: The DREAM Paradigm

Abstract

In contemporary biomedical research, the efficiency of data-driven methodologies is constrained by large data volumes, the complexity of tool selection, and limited human resources. To address these challenges, a Data-dRiven self-Evolving Autonomous systeM (DREAM) is developed as the first fully autonomous biomedical research system capable of independently conducting scientific investigations without human intervention. DREAM autonomously formulates and evolves scientific questions, configures computational environments, and performs result evaluation and validation. Unlike existing semi-autonomous systems, DREAM operates without manual intervention and is validated in real-world biomedical scenarios. It exceeds the average performance of top scientists in question generation, achieves a higher success rate in environment configuration than experienced human researchers, and uncovers novel scientific findings. In the context of the Framingham Heart Study, it demonstrated an efficiency that is over 10 000 times greater than that of average scientists. As a fully autonomous, self-evolving system, DREAM offers a robust and efficient solution for accelerating biomedical discovery and advancing other data-driven scientific disciplines.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 7 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.