autosynthesis-ab1d6303·1 events·first seen Aliases: AutoSynthesis
Researchers introduce AutoSynthesis, an end-to-end multi-agent system that automates the full meta-analysis pipeline: search strategy formulation, literature retrieval, study screening, eligibility assessment, data extraction, effect size computation, and random-effects meta-analysis. The system also supports heterogeneity analysis, risk-of-bias assessment, and produces PRISMA-aligned reports. In evaluation, AutoSynthesis screened over 28 studies and produced pooled effect estimates closely matching expert-conducted meta-analyses (Hedges' g). The work demonstrates that agentic AI can meaningfully scale quantitative evidence synthesis for science, medicine, and policy applications.