Event location in text: six recorded examples
The downloads contain six original inputs, their expected labels and one recorded API observation per input revision. Disagreements stay visible.
Provide the passage, specified event and candidate places. Separate where the event occurred from printing and archive locations, preserving multiple sites or an unresolved reference.
A passage can name the place where an event occurred, where a report was printed and where a copy was filed. The input names one event and three candidate places. This decision follows the relationship in the text instead of choosing the first place name.
Multiple preserves an event that actually spans several places, even if some are absent from the candidates. Two incompatible accounts of a single location instead need Unresolved. Outside means that the text gives one clear place but the supplied choices omit it. A reference such as here needs an explicit anchor in the original account.
The cited author describes asking Jev where archival animal mentions are located before a later reading stage. This text-only exercise supplies its own English passages, choices and fallback rules. It does not reproduce that pipeline, inspect original scans or confirm historical findings.
This original text exercise adapts the narrow location-screening question described in the cited author account. The English passages, candidate names and fallback rules are ours. It does not reproduce the upstream research pipeline, geocode places or verify historical events.

Use the sample to understand the rule, then test your own input for event location in text.

Provide the passage, specified event and candidate places.

Open the decision rules and keep the intended distinction between A, B, C, Multiple, Outside. Edit the choices only when testing your own rule.

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

Review uncertain or unexpected results before using a result. The label is a decision about the supplied information, not an executed action.
| Original input | Observed label | Leading probability |
|---|---|---|
| Event and printing location differ | A | 100% |
| Archive label differs from the event | B | 100% |
| A supplied anchor for here | C | 100% |
| One survey spans two sites | Multiple | 93.0% |
| Explicit place absent from the candidates | Outside | 100% |
| Here without an original reference point | Unresolved | 91.0% |
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. 6 of 6 results matched the authored fixture expectations.Verified
The downloads contain six original inputs, their expected labels and one recorded API observation per input revision. Disagreements stay visible.
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.
# Offline: inspect the downloaded saved observations. No API key is needed.
python preview_choice_workflow.py samples.json saved-api-results.json
# Optional live request: install the SDK and set TYPESAFE_API_KEY privately.
npm install @typesafe-ai/sdk@0.6.0
node run-choice-sample.mjs samples.json case-1
# The live response prints to the terminal; it does not update saved-api-results.json.The input contract and these limits define what the result means.
No. A dateline, printer's address or archive folder can name another place. Supply the event and the full passage so the decision can follow the stated relationship.
Only when the supplied text anchors it to the original speaker or writer's location. A later reprint's location does not supply that missing reference point.
Multiple is the expected label when one event explicitly spans several places. Unresolved preserves conflicting alternatives or missing context. The rule does not turn disagreement into evidence that the event happened in both places.
No. It selects from the place candidates you supply. A separate georeferencing process must resolve real-world identity, coordinates and uncertainty.
It describes the distribution over the supplied choices for this request. These six synthetic examples are functional tests with authored expectations, not a production accuracy benchmark.
No. The demo displays a label and distribution. The downloadable consumer produces a pending preview packet. Application code and a separate authorization policy would control any real action.
Edit the example above to see how a changed fact affects the decision, or open the Playground for another task.