What AI brings to research“RAG … grounds every answer in verifiable sources and mitigates the factual errors and hallucinations seen in standalone LLMs.” — Gavrilova & Galli (2026), Evidence-Based Dentistry
13,061 papers · continuously updated · last export: 4 Jul 2026livingmeta.ai

Meta-analysis overview

Full version

Available in the full version

Browsing this living review is free, but the meta-analysis overview — pooled effect sizes, forest plots and the per-paper effect table — is part of the full version.

What the full version shows here:

  • Random-effects pooling within each paradigm (never across paradigms)
  • Forest plots with heterogeneity (I², τ²) and a prediction interval
  • Per-paper effect sizes with r-conversion and 95% confidence intervals
  • LMQS quality badges per study