Triple

T23018082
Position Surface form Disambiguated ID Type / Status
Subject Terror by Night E573086 entity
Predicate producer P490 FINISHED
Object Howard Benedict 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: Howard Benedict | Statement: [Terror by Night, producer, Howard Benedict]
NED1 Entity disambiguation (via context triple) gpt-5-mini-2025-08-07
Target entity: Howard Benedict
Context triple: [Terror by Night, producer, Howard Benedict]
  • A. Howard Benedict chosen
    Howard Benedict was a film producer best known for his work on classic mid-20th-century movies, including entries in the Sherlock Holmes series.
  • B. Howard Benedict
    Howard Benedict is a character in the stage musical "Applause," serving as a powerful Broadway producer whose actions significantly influence the story’s backstage drama.
  • C. Henry Benenson
    Henry Benenson is an individual notable enough to be recognized as a prominent bearer of the Benenson surname.
  • D. William Budden
    William Budden was a 19th-century engineer known for his work on the Lancaster and Carlisle Railway in England.
  • E. John Honeyman
    John Honeyman was a prominent 19th-century Scottish architect known for designing notable public buildings and observatories.
  • 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_69e245b764cc8190a51be76f1d9611e1 completed April 17, 2026, 2:37 p.m.
NER Named-entity recognition batch_69f183e64a4c8190b8d29ed638c7fef8 completed April 29, 2026, 4:07 a.m.
Created at: April 17, 2026, 3:52 p.m.