Litmus NoteKnow before you commit
SearchAbout

Search

Two ways to search the literature

Find papers relevant to a topic whatever stance they take, or give a claim and see which papers agree, which disagree, and which stay unclear. In the second mode every stance is a draft backed by a quote from the abstract, and your browser checks that the quote is really there.

Finds work on your topic whatever stance it takes. With query drafting on, the model turns your phrase into several short queries and the records most of them agree on rank first.

Databases

Settings

Shown in the consent dialog and the disclosure statement.

Your own model key. It stays in this browser. Leave it empty to use the shared key this deployment already has, if there is one. The arguments mode uses it to draft queries and read each abstract against your claim.

Identifies your searches to OpenAlex. It goes out as the mailto parameter in each request URL.

Optional. A key raises the daily budget ten times, and searching works without one. Leave it empty to use the key this deployment already has.

Not used while Semantic Scholar is off. Optional: requests are limited to one per second, with backoff when the API pushes back. Leave it empty to use the key this deployment already has.

Search sends your query or claim, plus public paper metadata and abstracts, to a model run by someone else. Nothing else leaves this browser.

Provider: —
Endpoint: —
Model: —

Only public metadata and abstracts go out. Your consent is timestamped and stays in this browser. In the relevant-papers mode you can also cancel and switch query drafting off to search without the model at all.