Triple
T7310117
| Position | Surface form | Disambiguated ID | Type / Status |
|---|---|---|---|
| Subject | Islamic Republic |
E168068
|
entity |
| Predicate | mayIncludeOffice |
P11900
|
FINISHED |
| Object | supreme leader |
—
|
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: supreme leader | Statement: [Islamic Republic, mayIncludeOffice, supreme leader]
PD
Predicate disambiguation
gpt-5-mini-2025-08-07
Target predicate: mayIncludeOffice Context triple: [Islamic Republic, mayIncludeOffice, supreme leader]
-
A.
includedOffice
chosen
Indicates that one office is contained within, or forms part of, another office or organizational unit.
-
B.
usedByOffice
Indicates that something is utilized, operated, or employed by an office or office-related entity.
-
C.
officeContingentOn
Indicates that holding or obtaining one office, position, or role is dependent on the prior existence, status, or outcome of another specified office or condition.
-
D.
hasRelativeInOffice
Indicates that one entity has a family member who holds or held a particular office or official position.
-
E.
mayIncludeFeature
Indicates that one entity is allowed or able to contain, incorporate, or be associated with a particular feature.
- 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_69c6888d8e3c81909db79714903baf31 |
completed | March 27, 2026, 1:39 p.m. |
| NER | Named-entity recognition | batch_69c6ebdbd6f481908e69e53dab452656 |
completed | March 27, 2026, 8:43 p.m. |
| PD | Predicate disambiguation | batch_69c6e7705f4881909793071dee50c557 |
completed | March 27, 2026, 8:24 p.m. |
Created at: March 27, 2026, 3:01 p.m.