Triple
T28437325
| Position | Surface form | Disambiguated ID | Type / Status |
|---|---|---|---|
| Subject | State of California v. Robert Durst |
E715302
|
entity |
| Predicate | victimRelationshipToDefendant |
P101216
|
FINISHED |
| Object | longtime friend |
—
|
LITERAL 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: longtime friend | Statement: [State of California v. Robert Durst, victimRelationshipToDefendant, longtime friend]
PD
Predicate disambiguation
gpt-5-mini-2025-08-07
Target predicate: victimRelationshipToDefendant Context triple: [State of California v. Robert Durst, victimRelationshipToDefendant, longtime friend]
-
A.
victimRelationship
Indicates that one entity is the victim in relation to another entity involved in a harmful, criminal, or adverse act.
-
B.
victimRelation
Indicates that one entity is the victim or target of harm, wrongdoing, or an adverse action caused by another entity.
-
C.
hasRelationshipToPerpetrator
chosen
Indicates that an entity has a specified type of relationship or connection to the perpetrator of an act or event.
-
D.
victimParty
Indicates that one party is the victim or harmed side in an event, action, or situation involving another party.
-
E.
victimRole
Indicates that one entity participates in an event or situation specifically in the role of the victim or harmed party.
- 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_69efd6b253888190b3c7222ed6a403a8 |
completed | April 27, 2026, 9:35 p.m. |
| NER | Named-entity recognition | batch_69f6645ba71c81908044ade6ab577018 |
completed | May 2, 2026, 8:53 p.m. |
| PD | Predicate disambiguation | batch_69f663362c008190a22afed262f1e426 |
completed | May 2, 2026, 8:48 p.m. |
Created at: April 28, 2026, 1:43 a.m.