Triple
T16766252
| Position | Surface form | Disambiguated ID | Type / Status |
|---|---|---|---|
| Subject | Dragonheart |
E407471
|
entity |
| Predicate | director |
P255
|
FINISHED |
| Object | Rob Cohen |
—
|
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: Rob Cohen | Statement: [Dragonheart, director, Rob Cohen]
NED1
Entity disambiguation (via context triple)
gpt-5-mini-2025-08-07
Target entity: Rob Cohen Context triple: [Dragonheart, director, Rob Cohen]
-
A.
Rob Cohen
chosen
Rob Cohen is an American film director and producer best known for helming high-profile action and adventure movies such as "The Fast and the Furious" and "xXx."
-
B.
Jon Cohen
Jon Cohen is a screenwriter best known for co-writing the science fiction film "Minority Report."
-
C.
Robert Cohen
Robert Cohen is a cellist recognized as one of the distinguished pupils of renowned cello teacher William Pleeth.
-
D.
Robert Cohen
Robert Cohen is an audio engineer known for his technical work on music recordings, including serving in engineering roles on major studio projects.
-
E.
Jake Cohen
Jake Cohen is known primarily as the son of Michael Cohen, the former personal attorney to Donald Trump.
- 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_69d8839174188190909f190097207065 |
completed | April 10, 2026, 4:58 a.m. |
| NER | Named-entity recognition | batch_69e3b0330d6081908ce99f14c70b90f2 |
completed | April 18, 2026, 4:24 p.m. |
Created at: April 10, 2026, 5:21 a.m.