Small decisions. Interesting possibilities.Submit contentSubmit

How it uses Jev

When explicitly configured and enabled, the optional extension sends a paper title and abstract plus one Choice question per configured necessary inclusion condition. Each question offers meets, violates, and insufficient. The output is a review-required proposal, not an automatic ledger decision.

What Jev decides

Illustrative title and abstract inclusion-condition requestExample answers · not a recorded Jev response · Source ↗
Question 1 · population
YOUR APP
INSTRUCTION

Only from title and abstract, judge whether the material satisfies this necessary inclusion condition. Missing description is not a violation; choose insufficient for complex inference or missing evidence.

STATE

The paper title and abstract, plus configured named atomic inclusion conditions. The included config example names population, empirical method, and outcome conditions.

JEV · CHOICE
  1. meets
  2. violates
  3. insufficient

App workflow

  1. Prepare a bounded paper record

    Read one paper title and abstract from the frozen research task. The extension allowlists these fields for the request.

  2. Ask Jev per condition

    Build one Choice question per configured condition, then make a budgeted request or reuse a cached response.

  3. Create review proposal

    Apply the configured probability threshold. Mark the paper as include, exclude, or uncertain, and route every proposal for evidence verification or further review.

Why it is interesting

It decomposes a research question into named atomic conditions, preserves uncertainty, and keeps machine screening proposals separate from the evidence-backed human research ledger.