Triple
T23555995
| Position | Surface form | Disambiguated ID | Type / Status |
|---|---|---|---|
| Subject | Poncirus |
E578187
|
entity |
| Predicate | relatedTo |
P37
|
FINISHED |
| Object | Citrus |
—
|
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: Citrus | Statement: [Poncirus, relatedTo, Citrus]
NED1
Entity disambiguation (via context triple)
gpt-5-mini-2025-08-07
Target entity: Citrus Context triple: [Poncirus, relatedTo, Citrus]
-
A.
Citrus
chosen
Citrus is a genus of flowering trees and shrubs known for producing popular aromatic fruits such as oranges, lemons, limes, and tangerines.
-
B.
Cítricos
Cítricos is an upscale, Mediterranean-inspired restaurant located at Disney’s Grand Floridian Resort & Spa in Walt Disney World.
-
C.
Citrus sinensis
Citrus sinensis is the sweet orange tree, a widely cultivated citrus species known for its juicy, sweet-tasting fruits and global importance as a fresh fruit and juice crop.
-
D.
Zitrone
Zitrone is a French surname most notably borne by Léon Zitrone, a prominent 20th-century television journalist and presenter in France.
-
E.
Citrus reticulata
Citrus reticulata is the mandarin orange, a small, sweet, and easy-to-peel citrus fruit widely cultivated and consumed around the world.
- F. None of above.
- G. Unsure - the case is ambiguous/there is not enough information to decide.
Provenance (2 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_69e245fa93448190919cb04534560542 |
completed | April 17, 2026, 2:38 p.m. |
| NER | Named-entity recognition | batch_69f1aed37f248190992f040f4d22f0bf |
completed | April 29, 2026, 7:10 a.m. |
Created at: April 17, 2026, 6:12 p.m.