Triple
T6865112
| Position | Surface form | Disambiguated ID | Type / Status |
|---|---|---|---|
| Subject | Lockheed U-2 |
E158380
|
entity |
| Predicate | hasVariant |
P455
|
FINISHED |
| Object | TR-1 |
E158381
|
NE 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: TR-1 | Statement: [Lockheed U-2, hasVariant, TR-1]
NED1
Entity disambiguation (via context triple)
gpt-5-mini-2025-08-07
Target entity: TR-1 Context triple: [Lockheed U-2, hasVariant, TR-1]
-
A.
TR-1A
chosen
The TR-1A is a high-altitude tactical reconnaissance aircraft developed from the Lockheed U-2, optimized for battlefield surveillance and intelligence-gathering missions.
-
B.
TR-25
TR-25 is the statistical and administrative region code assigned to Turkey’s Erzurum Province.
-
C.
TR-85
The TR-85 is a Romanian main battle tank developed during the Cold War as an improved, domestically produced evolution of the Soviet T-55 design.
-
D.
TR-42
TR-42 is the ISO 3166-2 regional code assigned to Turkey’s Konya Province.
-
E.
RTR
RTR is a Swiss public broadcasting division that produces and distributes radio, television, and online content in the Romansh language.
- F. None of above.
- G. Unsure - the case is ambiguous/there is not enough information to decide.
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_69c68831e3648190a643c328122e4d43 |
completed | March 27, 2026, 1:37 p.m. |
| NER | Named-entity recognition | batch_69c6d88af6d88190ac9faa32fa1bfa0e |
completed | March 27, 2026, 7:20 p.m. |
| NED1 | Entity disambiguation (via context triple) | batch_69c72ff153d48190a4b0d4e403457fe8 |
completed | March 28, 2026, 1:33 a.m. |
Created at: March 27, 2026, 2:21 p.m.