Triple

T20154114
Position Surface form Disambiguated ID Type / Status
Subject Room at the Top E491509 entity
Predicate producer P490 FINISHED
Object John Woolf 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: John Woolf | Statement: [Room at the Top, producer, John Woolf]
NED1 Entity disambiguation (via context triple) gpt-5-mini-2025-08-07
Target entity: John Woolf
Context triple: [Room at the Top, producer, John Woolf]
  • A. John Woolf chosen
    John Woolf was a prominent British film producer known for backing acclaimed mid-20th-century films, including several major literary adaptations and award-winning dramas.
  • B. Harry Kenneth Woolf
    Harry Kenneth Woolf, known as Lord Woolf, is a prominent British jurist who served as Lord Chief Justice of England and Wales and is noted for major reforms to the civil justice system.
  • C. Wolfgang Wells
    Wolfgang Wells is one of the many alternate-universe versions of genius scientist Harrison Wells, portrayed by Tom Cavanagh in the TV series "The Flash."
  • D. Essex Wright
    Essex Wright is the son of Chicago rapper G Herbo.
  • E. Eliot Wald
    Eliot Wald was an American comedy writer and screenwriter known for his work on films and television, including contributing to the script of the submarine comedy "Down Periscope."
  • 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_69da6265f8f0819080b29c752a574088 completed April 11, 2026, 3:01 p.m.
NER Named-entity recognition batch_69e667de9bec8190836887c86dbcf28d completed April 20, 2026, 5:52 p.m.
Created at: April 11, 2026, 11:34 p.m.