Triple
T31431364
| Position | Surface form | Disambiguated ID | Type / Status |
|---|---|---|---|
| Subject | Atlas Cheetah |
E801805
|
entity |
| Predicate | upgradeTarget |
P17584
|
FINISHED |
| Object | Dassault Mirage III |
—
|
NE NERFINISHED |
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: Dassault Mirage III | Statement: [Atlas Cheetah, upgradeTarget, Dassault Mirage III]
PD
Predicate disambiguation
gpt-5-mini-2025-08-07
Target predicate: upgradeTarget Context triple: [Atlas Cheetah, upgradeTarget, Dassault Mirage III]
-
A.
upgradeGoal
Indicates that an entity has a target or intended improvement to a higher level, version, or state relative to its current condition.
-
B.
upgradeTargetUnits
Indicates that certain units are designated to be improved or converted to a higher or more advanced version.
-
C.
upgradeType
Indicates the specific kind or category of improvement or enhancement applied to an entity relative to its previous state.
-
D.
upgradePath
chosen
Indicates the relationship by which one entity can be improved, enhanced, or transitioned into another, typically more advanced, version or state.
-
E.
upgradeFor
Indicates that one entity serves as an improved or enhanced replacement for 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_69f348c475348190bf579ca858eec77c |
completed | April 30, 2026, 12:19 p.m. |
| NER | Named-entity recognition | batch_69f6a5f71b2c8190aade8a83f465be0c |
completed | May 3, 2026, 1:33 a.m. |
| PD | Predicate disambiguation | batch_69f69fe66df08190958558d63ee623d9 |
completed | May 3, 2026, 1:07 a.m. |
Created at: April 30, 2026, 8:57 p.m.