Agents Lookup

Use cases / Decision models (Jev)

Sort a large set of papers or articles into topics

Give Jev a fixed topic list and a title plus abstract for each paper, then browse or publish the sorted collection.

DifficultyBeginner
Setup time1-2 hours
CategoryDecision models (Jev)

Author's figure: Author's figure: 1,018 papers for $0.08 with 256 ms median latency (Hassan). Reported by the builder, not measured by us.

The use case

Research teams, analysts and newsletter writers face thousands of papers or reports and need them grouped before anyone reads. Each title and abstract goes to Jev with one Choice question over a topic list you control. The sorted set can be published as a browsable site or used to pick reading. Two rebuilds checked Jev's labels with a larger model as a grader, which is a sensible habit. A person should spot-check each topic, merge or split categories that do not work, and treat the sort as a reading aid rather than a literature review.

Example workflow

  1. Collect titles and abstracts from arXiv or a reference manager
  2. Write a topic list of 10-30 categories with one-line descriptions
  3. Ask Jev a Choice question per paper
  4. Grade a sample of labels with a larger LLM or by hand
  5. Publish or share the sorted list in Notion or a static site

Tools

Risks and guardrails

Sources

Written by Agents Lookup. Credits and official pages for the tools this use case names: