Empowering AI data scientists using a multi-agent LLM framework with self-evolving capabilities for autonomous, tool-aware biomedical data analyses - PubMed
7 hours ago
- #AI in Biomedicine
- #Bioinformatics Tools
- #LLM Agents
- BioMedAgent is a self-evolving multi-agent LLM framework designed for biomedical data analysis.
- It learns to use and chain bioinformatics tools into workflows through interactive exploration and memory retrieval.
- The system allows biomedical users to initiate tasks using natural language without requiring computational expertise.
- Evaluated on the BioMed-AQA benchmark (327 tasks), BioMedAgent achieved a 77% success rate, surpassing other LLM agents.
- It demonstrated robust generalization on the external BixBench dataset.
- Applications include autonomous cross-omics analysis, machine learning modeling, and pathology image segmentation.
- The framework has potential to advance biomedical research and extend to other scientific domains with complex tool integration.