Triple
T16072422
| Position | Surface form | Disambiguated ID | Type / Status |
|---|---|---|---|
| Subject | John Frankenheimer |
E389895
|
entity |
| Predicate | notableWork |
P4
|
FINISHED |
| Object | Grand Prix |
E12909
|
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: Grand Prix | Statement: [John Frankenheimer, notableWork, Grand Prix]
NED1
Entity disambiguation (via context triple)
gpt-5-mini-2025-08-07
Target entity: Grand Prix Context triple: [John Frankenheimer, notableWork, Grand Prix]
-
A.
Grand Prix
The Grand Prix is the top prize awarded to the overall winner of the Sopot International Song Festival, recognizing the festival’s most outstanding performance.
-
B.
Grand Prix
chosen
The Grand Prix is one of the most prestigious awards at the Cannes Film Festival, typically regarded as the festival’s second-highest honor after the Palme d’Or.
-
C.
Grand Prix
Grand Prix is a classic 1982 Activision racing video game for the Atari 2600 known for its colorful graphics and smooth side-scrolling gameplay.
-
D.
Grand Prix
The Grand Prix is the top overall prize awarded at the Rovinj Photodays photography festival, recognizing the most outstanding work in the competition.
-
E.
Grand Prix
Grand Prix is a primary racing mode in the Mario Kart series where players compete in a series of cup races against computer-controlled opponents.
- 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_69d86daf32ec8190a8c0466c8f49c3c0 |
completed | April 10, 2026, 3:25 a.m. |
| NER | Named-entity recognition | batch_69e183bf6c488190b0099a00f13f2a69 |
completed | April 17, 2026, 12:50 a.m. |
| NED1 | Entity disambiguation (via context triple) | batch_69ffe484cef08190a3797c91a7025081 |
completed | May 10, 2026, 1:51 a.m. |
Created at: April 10, 2026, 4:57 a.m.