LivingMeta AI in Education

Every paper is independently analyzed by six AI personas, each with a different cognitive focus (methodologist, skeptic, synthesizer, etc.). They extract the same 75 fields, and a consensus algorithm resolves disagreements field-by-field. Every factual claim must include a verbatim quote from the source paper — extractions without textual evidence are automatically flagged. Agreement scores are visible on each paper card.

The agenda is built automatically from the evidence in this instance. First, AI extractions identify research gaps across all papers (only highly relevant papers, scored 7+ out of 10, contribute). These gaps are clustered into themes using TF-IDF text analysis. Each theme is then scored on five weighted dimensions: how frequently the gap appears, the quality of the source papers, feasibility of addressing it, recency of the evidence, and how broadly across the field the gap surfaces. An AI then reviews the top-ranked themes and filters out anything too narrow or niche — only priorities relevant to the broader field survive. The result is a ranked list that updates with every new paper added to the platform.

Yes — The Lab provides AI-coached research guidance through 8 phases, from exploring the literature to writing your manuscript. Select the 'Junior Researcher' role when starting a thread, and the system will scaffold your process with Socratic questions. You can export papers to CSV or BibTeX at any time.

Those tools address individual stages — Elicit does per-paper extraction, Consensus searches abstracts, Cochrane produces static reviews. LivingMeta integrates extraction, gap analysis, coached investigation, and resource discovery into a single continuously updated workflow. The platform is the methodology.

Start a thread in The Lab describing what you found. The instance owner reviews submissions and can trigger corrections or additions. Papers flow through the same classification and extraction pipeline as all other content.