Triple
T9129451
| Position | Surface form | Disambiguated ID | Type / Status |
|---|---|---|---|
| Subject | JFKSWCS |
E219044
|
entity |
| Predicate | languageTraining |
P28135
|
FINISHED |
| Object | foreign language instruction for Special Forces soldiers |
—
|
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: foreign language instruction for Special Forces soldiers | Statement: [JFKSWCS, languageTraining, foreign language instruction for Special Forces soldiers]
PD
Predicate disambiguation
gpt-5-mini-2025-08-07
Target predicate: languageTraining Context triple: [JFKSWCS, languageTraining, foreign language instruction for Special Forces soldiers]
-
A.
languageOfTeachings
Indicates the language in which teachings, lessons, or instructional content are delivered or expressed.
-
B.
languageCapacity
Indicates the extent to which an entity is able to understand, produce, or otherwise use language.
-
C.
languagesSpoken
Indicates that an entity is able to communicate using one or more specified languages.
-
D.
trainingDomain
chosen
Indicates that an entity is associated with or operates within a particular field, area, or domain of training.
-
E.
languageRevivalMethod
Indicates the method or strategy used to revive or revitalize a language that is endangered, dormant, or no longer actively spoken.
- 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_69ca83debfc0819095800583e97ab10f |
completed | March 30, 2026, 2:08 p.m. |
| NER | Named-entity recognition | batch_69cca8cc85b081908ee80db0e3b7df15 |
completed | April 1, 2026, 5:10 a.m. |
| PD | Predicate disambiguation | batch_69cc6601d77881908299d58db6e64937 |
completed | April 1, 2026, 12:25 a.m. |
Created at: March 30, 2026, 7:18 p.m.