Triple

T9117224
Position Surface form Disambiguated ID Type / Status
Subject Pretty Woman E218750 entity
Predicate director P255 FINISHED
Object Garry Marshall E90787 NE 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: Garry Marshall | Statement: [Pretty Woman, director, Garry Marshall]
NED1 Entity disambiguation (via context triple) gpt-5-mini-2025-08-07
Target entity: Garry Marshall
Context triple: [Pretty Woman, director, Garry Marshall]
  • A. Garry Marshall chosen
    Garry Marshall was an American filmmaker, producer, and actor best known for creating and directing popular television sitcoms and romantic comedy films.
  • B. Barry Hyman
    Barry Hyman is one of the children of American literary critic and professor Stanley Edgar Hyman.
  • C. Don Hahn
    Don Hahn is an American film producer best known for overseeing several of Disney’s most acclaimed animated features, including Beauty and the Beast and The Lion King.
  • D. Jerry Zucker
    Jerry Zucker is an American film director, producer, and screenwriter best known as part of the Zucker, Abrahams and Zucker team behind comedies like "Airplane!" and for directing the romantic fantasy drama "Ghost."
  • E. Karey Kirkpatrick
    Karey Kirkpatrick is an American screenwriter and director known for his work on animated and family films such as Chicken Run, Over the Hedge, and Smallfoot.
  • F. None of above.
  • G. Unsure - the case is ambiguous/there is not enough information to decide.

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_69ca83dc94ac8190b9ef42684d36ff39 completed March 30, 2026, 2:08 p.m.
NER Named-entity recognition batch_69cca8a5e2ac8190b602ef0c77deb2fa completed April 1, 2026, 5:09 a.m.
NED1 Entity disambiguation (via context triple) batch_69d0307299ec8190acade4f388642e23 completed April 3, 2026, 9:26 p.m.
Created at: March 30, 2026, 7:17 p.m.