Triple
T12188569
| Position | Surface form | Disambiguated ID | Type / Status |
|---|---|---|---|
| Subject | Operation Castle |
E290400
|
entity |
| Predicate | testedDeviceType |
P104198
|
FINISHED |
| Object | two-stage thermonuclear device |
—
|
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: two-stage thermonuclear device | Statement: [Operation Castle, testedDeviceType, two-stage thermonuclear device]
PD
Predicate disambiguation
gpt-5-mini-2025-08-07
Target predicate: testedDeviceType Context triple: [Operation Castle, testedDeviceType, two-stage thermonuclear device]
-
A.
testedDevice
Indicates that one entity has performed a test on, or experimentally evaluated, another entity as a device.
-
B.
targetedDevice
Indicates that one entity is the specific device toward which another entity’s action, effect, or configuration is directed.
-
C.
controlsDeviceType
Indicates that an entity has authority over or can operate a specific type or category of device.
-
D.
connectsDeviceType
Indicates a relationship where one entity is associated with or linked to a specific type or category of device.
-
E.
definesDeviceType
Indicates that one entity specifies or assigns the device type classification of another entity.
- F. None of above. chosen
Provenance (4 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_69d6ab64de5881908d56eb7a75c6cc69 |
completed | April 8, 2026, 7:24 p.m. |
| NER | Named-entity recognition | batch_69d938cd2edc8190b1971349dbc0dee0 |
completed | April 10, 2026, 5:52 p.m. |
| PD | Predicate disambiguation | batch_69d91c38321c819080d500d0d64a04f6 |
completed | April 10, 2026, 3:50 p.m. |
| PDg | Predicate description generation | batch_69d938ca32908190bd56f563efcbe8a0 |
completed | April 10, 2026, 5:52 p.m. |
Created at: April 8, 2026, 9:50 p.m.