Triple
T20098766
| Position | Surface form | Disambiguated ID | Type / Status |
|---|---|---|---|
| Subject | Fatu Hiva |
E496476
|
entity |
| Predicate | hasHighestPoint |
P210
|
FINISHED |
| Object | Mount Touaouoho |
—
|
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: Mount Touaouoho | Statement: [Fatu Hiva, hasHighestPoint, Mount Touaouoho]
NED1
Entity disambiguation (via context triple)
gpt-5-mini-2025-08-07
Target entity: Mount Touaouoho Context triple: [Fatu Hiva, hasHighestPoint, Mount Touaouoho]
-
A.
Mont Touaouoho
chosen
Mont Touaouoho is the principal mountain peak on the remote Polynesian island of Fatu Hiva in the Marquesas Islands of French Polynesia.
-
B.
Mount Phousi
Mount Phousi is a prominent hill in the center of Luang Prabang, Laos, known for its Buddhist shrines and panoramic views of the city and Mekong River.
-
C.
Mount Loilaeng
Mount Loilaeng is a prominent peak in eastern Myanmar, recognized as the highest mountain in the Shan Hills range.
-
D.
Mount Vineuo
Mount Vineuo is the tallest mountain on the D'Entrecasteaux Islands of Papua New Guinea, known for its prominent peak and rugged terrain.
-
E.
Mount Heha
Mount Heha is the tallest mountain in Burundi, located in the Burundi Highlands near the city of Bujumbura.
- 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_69da626eee3881909f3454986d4a6511 |
completed | April 11, 2026, 3:02 p.m. |
| NER | Named-entity recognition | batch_69e6666e306c81909c0ef617e0f6fccf |
completed | April 20, 2026, 5:46 p.m. |
Created at: April 11, 2026, 11:25 p.m.