Jev contrast function

Review what a contrast adds to a sentence

Paste the contrast and its context. Distinguish a label swap from a concrete difference or evidence-based correction without treating the wording as proof of truth.

What does the contrast do?Live API
Try an example

The second label does not explain a new function.

Source: API-collected workflow inspiration; original demonstration below

Source and example scope

The API-collected roundup links Sniff Test. Its pinned rules define the text distinction adapted here. This original Choice demonstration uses synthetic inputs and adds explicit missing-context handling; the source linter uses its own yes-or-no judgments. It does not reproduce that application or validate the truth of a claim.

Classify the contrast.

Decision rules and labels
Relabeling onlyEvidence-based correctionInformative contrastNo replacement contrastNeed context
A real response from the Jev API

Edit any input to explore a different decision. No request is sent until you run it.

Your next decision

Not run yet

Choose an example or write your own input, then run it to see the ranked choices.

Results are not stored by this site. Your submitted text is processed by the API provider.

The decision

A contrast can correct a fact, explain a reason or merely change a label.

Calling a dashboard a command center supplies a different description without explaining what changes. The schedule preset adds a practical reason for publishing on Tuesday. Those passages deserve different review notes even though both contain a contrast.

The row-count preset includes its own synthetic manifest. The model can identify that the sentence uses evidence to correct a number. It cannot authenticate the manifest or establish that the real export contains 480 rows.

Use the result as an editorial observation. The page does not rewrite the paragraph or automatically delete contrasting phrases; context and house style still determine the edit.

The API-collected roundup links Sniff Test. Its pinned rules define the text distinction adapted here. This original Choice demonstration uses synthetic inputs and adds explicit missing-context handling; the source linter uses its own yes-or-no judgments. It does not reproduce that application or validate the truth of a claim.

Illustration of selecting a caption token for emphasis in an edit timeline
Illustration of the decision flow.
How it works

Try the example in four steps

Use the sample to understand the rule, then test your own input for contrast in prose.

  1. Illustration of adding the text or evidence for a decision

    Add the information to check

    Paste the contrast and its context.

  2. Illustration of defining a question and its allowed choices

    Confirm the question and labels

    Open the decision rules and keep the intended distinction between Relabeling only, Evidence-based correction, Informative contrast, No replacement contrast, Need context. Edit the choices only when testing your own rule.

  3. Illustration of sending one input and comparing the returned choices

    Run the decision

    Use a sample first, then try your own input. Compare the selected label with the other returned probabilities.

  4. Illustration of reviewing a result and an uncertain item before using it

    Review the next step

    Review Need context before using a result. The label is a decision about the supplied information, not an executed action.

Our API test

Recorded API observations

Recorded API observations
Original inputObserved labelLeading probability
Label swapRelabeling only99.0%
Correction with a recordEvidence-based correction100%
Schedule and reasonInformative contrast85.0%
Plain negationNo replacement contrast76.0%
Separate factEvidence-based correction72.0%
Truncated comparisonNeed context100%

These are six original synthetic inputs, tested once per request revision. The percentage is the leading option probability, not measured accuracy. Authored expectations and any disagreements remain in the downloadable evidence. 5 of 6 results matched the authored fixture expectations. For Separate fact, the expectation was Informative contrast; the observed choice was Evidence-based correction.Verified

Understand the results

Replay the saved synthetic requests to inspect each recorded decision

Replay the saved synthetic requests to inspect each recorded decision. Authored expectations and model disagreements remain available; these examples are not a production accuracy benchmark.

Using the result in your application

The single-request SDK runner uses your private TYPESAFE_API_KEY. The offline consumer reads saved observations, checks request hashes and prints a pending output. It does not call the model or change any external system.

Runnable offline example

Run one sample, then inspect the local packet

View the code example
Single-request replay
npm install @typesafe-ai/sdk@0.6.0
# Set TYPESAFE_API_KEY privately in your shell.
node run-choice-sample.mjs samples.json case-1
python preview_choice_workflow.py samples.json saved-api-results.json
Scope of this demo

Where this decision fits

Useful scope

  • Reviewing repeated rhetorical label swaps
  • Keeping useful corrections out of a style-warning queue
  • Explaining why a contrast contributes information

Outside this demonstration

  • Authenticating a cited record
  • Automatically deleting contrasting sentences
  • Judging an incomplete quotation
Example FAQ

Questions about contrast function.

Input scope and downstream handling determine what this judgment means.

Are all contrasts a writing problem?

No. The supplied rule separates corrections and added information from a label swap that contributes nothing else.

Does Evidence-based correction mean the correction is true?

It means supporting evidence is present in the supplied passage. Its authenticity remains a separate check.

Is the leading probability an accuracy measurement?

No. It is one value in the returned distribution over this request’s choices. The six synthetic fixtures have authored expectations; they do not measure production accuracy.

What happens on an API failure?

The window displays an error without a fresh judgment. An incomplete-input outcome describes the supplied evidence and is not used as a fallback for provider failure.

Try your own contrast function input

Edit the input above, or open the Playground to explore another decision.