Triple
T24717290
| Position | Surface form | Disambiguated ID | Type / Status |
|---|---|---|---|
| Subject | Flag of Paraguay |
E612196
|
entity |
| Predicate | reverseSealFeatures |
P157197
|
FINISHED |
| Object | staff |
—
|
LITERAL FINISHED |
How this triple was built (2 steps)
Every LLM step that produced this triple, in pipeline order — named-entity classification, the disambiguation choices (the exact options shown, with the pick highlighted), and the generated description. The batch + timestamp of each is in the Provenance table below.
NER
Named-entity recognition
gpt-5-mini
Instruction
Given a phrase, classify it is english named entity (e.g., persons, organizations, works of art) in Latin script, or not (e.g., literals, dates, URLs, verbose phrases). For disambiguation, the statement where the phrase occurs as object is also given. Please return a JSON object with `phrase` (string, the phrase being analyzed) and `is_ne` (boolean, indicating whether the phrase is a Named Entity).
Input
Phrase: staff | Statement: [Flag of Paraguay, reverseSealFeatures, staff]
PD
Predicate disambiguation
gpt-5-mini-2025-08-07
Target predicate: reverseSealFeatures Context triple: [Flag of Paraguay, reverseSealFeatures, staff]
-
A.
reverseSealFeatures
chosen
Indicates that the direction or orientation of seal-related features is inverted or swapped relative to a reference configuration.
-
B.
reverseFeature
Indicates that one feature is the inverse or opposite counterpart of another feature in a given context.
-
C.
reverseInscription
Indicates that one entity is inscribed as the reverse or mirror image of another entity’s inscription.
-
D.
secondSealAssociatedWith
Indicates an associative relationship in which something is linked or connected specifically to a second seal.
-
E.
usedSeal
Indicates that an entity has applied or employed a seal (e.g., for closure, authentication, or protection) on another entity.
- F. None of above.
Provenance (3 batches)
The batch behind each pipeline step, in order, with when it ran. Timestamps are batch-level — stages were processed in waves, so the object chain (NER → NED1 → NEDg → NED2) reads in order, but predicate / elicitation batches can sit in a different wave.
| Step | Stage | Batch ID | Status | When |
|---|---|---|---|---|
| creating | Elicitation | batch_69e2d7d6e7a48190bb43b0d8bb1137a0 |
completed | April 18, 2026, 1:01 a.m. |
| NER | Named-entity recognition | batch_69f453035f508190be83a3d521723acf |
completed | May 1, 2026, 7:15 a.m. |
| PD | Predicate disambiguation | batch_69f44d6ef33081908f5d36ba1ae5f473 |
completed | May 1, 2026, 6:51 a.m. |
Created at: April 18, 2026, 3:38 a.m.