Triple

T1749735
Position Surface form Disambiguated ID Type / Status
Subject Jason Taylor E38411 entity
Predicate worksAs P32113 FINISHED
Object head of Christian Grey's personal security team 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: head of Christian Grey's personal security team | Statement: [Jason Taylor, worksAs, head of Christian Grey's personal security team]
PD Predicate disambiguation gpt-5-mini-2025-08-07
Target predicate: worksAs
Context triple: [Jason Taylor, worksAs, head of Christian Grey's personal security team]
  • A. worksThrough
    Indicates that one entity performs an action, causes an effect, or achieves a result by using, employing, or acting via another entity as an intermediary or means.
  • B. worksTo
    Indicates that one entity performs work or exerts effort in order to achieve, support, or contribute to another entity or outcome.
  • C. worksFor
    Indicates that one entity is employed by or performs work on behalf of another entity, typically an organization or individual.
  • D. worksOnProgram
    Indicates that an entity is actively involved in contributing effort or performing tasks on a particular program.
  • E. worksOnIssue
    Indicates that an entity (typically a person or team) is actively engaged in addressing, resolving, or contributing work toward a specific issue.
  • F. None of above. chosen

Provenance (4 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_69a8862bdb2081908aefe831c8aa8017 completed March 4, 2026, 7:21 p.m.
NER Named-entity recognition batch_69ab630e7d008190a8c673665d9672bb completed March 6, 2026, 11:28 p.m.
PD Predicate disambiguation batch_69aa61c5a18481909bc49e0c54d64314 completed March 6, 2026, 5:10 a.m.
PDg Predicate description generation batch_69ab630cb34881908c9fb7ed5dedcd77 completed March 6, 2026, 11:28 p.m.
Created at: March 4, 2026, 7:31 p.m.