Triple
T33969062
| Position | Surface form | Disambiguated ID | Type / Status |
|---|---|---|---|
| Subject | Carl (Van Helsing) |
E870938
|
entity |
| Predicate | helpsFight |
P129918
|
FINISHED |
| Object | Dracula |
—
|
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: Dracula | Statement: [Carl (Van Helsing), helpsFight, Dracula]
PD
Predicate disambiguation
gpt-5-mini-2025-08-07
Target predicate: helpsFight Context triple: [Carl (Van Helsing), helpsFight, Dracula]
-
A.
diesFighting
Indicates that an entity dies as a direct result of engaging in a fight or combat.
-
B.
helpsIn
chosen
Indicates that one entity provides assistance, support, or aid to another entity in performing or achieving a particular task, activity, or goal.
-
C.
fightingOutOf
Indicates that an individual is competing or representing themselves in a fight or match while being officially associated with a particular location, camp, or organization.
-
D.
armedAgainst
Indicates that one entity is equipped with weapons or defensive capabilities specifically in preparation for potential conflict with another entity.
-
E.
fightsUsing
Indicates that one entity engages in combat or conflict by employing another entity as its weapon, tool, or method of fighting.
- F. None of above.
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_69f3499ce8e88190b66e1d49ad8c7037 |
completed | April 30, 2026, 12:22 p.m. |
| NER | Named-entity recognition | batch_69f70b966860819089cf92927f47c5f1 |
completed | May 3, 2026, 8:47 a.m. |
| PD | Predicate disambiguation | batch_69f70abe43e08190b2a30930d96247c1 |
completed | May 3, 2026, 8:43 a.m. |
Created at: May 1, 2026, 1:50 a.m.