Give it a text, and if you like your reading of it. Sync-Turing returns what the text says, each point tied to the text's own words, what it does not say and where it pulls two ways, and checks your reading sentence by sentence: is it in sync with the text?
Today's capacity for the research tools has been used. It opens again at midnight, Paris time.
Sign in with your email to use your three readings a day. We send a link that works once; there is no password.
Kept until you delete them or your account.
What the text says. Three to eight points, each with one or two exact quotes from your text, shown in the sentence they stand in. Each point is marked as something the text states, or as a reading of it.
How it argues, what it does not say, where it holds tension. Its definitions and the steps of its argument; what a reader might assume and the text does not state; the places where it pulls two ways, both sides quoted.
Your reading, sentence by sentence. Each of your sentences is supported (with the quote), supported elsewhere (the idea is in the text, but not as you put it), held both ways (the text says it and something that pulls against it), the text says otherwise (with the words that say so), or not in the text (it may be your own idea, which is no fault, but it is not the text's). A result the tool cannot confirm against the text is shown as such.
Sync. How your sentences fall into those results, and how many points of the text your reading does not touch. It measures whether your sentences agree with the text, not whether your reading is complete.
Autonomous agents and scripts use the same tool through an HTTPS API, with the same three readings a day per key. A machine can submit a document and its own summary, and get back which of its sentences the document bears out. A key belongs to the person who signs in here; creating a new key revokes the previous one.
curl -s https://research-api.llulllab.com/research/v1/sync-turing/ask \
-H "Authorization: Bearer $LLULL_RESEARCH_KEY" \
-H "Content-Type: application/json" \
-H "Idempotency-Key: <a unique id per reading>" \
-d '{"text": "... 200 to 40,000 characters ...", "reading": "... optional, up to 3,000 characters and 40 sentences ..."}'
# 202 {"job_id": "...", "status": "queued", "poll": "https://research-api.llulllab.com/research/v1/job/<job_id>"}
curl -s https://research-api.llulllab.com/research/v1/job/<job_id> \
-H "Authorization: Bearer $LLULL_RESEARCH_KEY"
# status: queued | running | done | refused | out_of_scope | failed
# result: {"points": [{"claim", "status": "stated|interpretation", "quotes": [{"quote", "context"}], "in_reading"}],
# "argument", "definitions", "distinctions", "not_said", "tensions", "open_questions",
# "alignment": [{"n", "sentence", "result": "supported|elsewhere|both|contradicted|not_in_text|question|not_confirmed", "quotes", "note"}],
# "sync": {"sentences", "supported", "elsewhere", "both", "contradicted", "not_in_text", "not_confirmed", "questions",
# "points_of_text", "points_not_in_reading"},
# "in_short", "language", "ai_generated": true, "generator", "notice"}
A reading takes one to three minutes; poll every few seconds. A failed reading does not use one of your three, and a repeated request with the same Idempotency-Key returns the same job. Answers are generated by a computational system: if you pass them on to people, say so.
Sync-Turing is Llull Lab's method: an evaluation method measuring the ability of humans and machines to use their communication capacities at maximum efficiency and in high synchronization, reaching a shared rhythm between two distinct entities. This tool applies it to reading: it measures how far a reader, human or machine, and a text are in sync, and gives back a reading of the text that you can check against the text yourself. How the tool works is not disclosed.
Research and study use. The reading is computational: check what matters to you against the text. Not legal or other professional advice. Sync-Turing is a service by Llull Lab (Firat Ilim, entrepreneur individuel), Paris.