Triple

T22962041
Position Surface form Disambiguated ID Type / Status
Subject Intruders E570926 entity
Predicate castMember P1668 FINISHED
Object Carice van Houten 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: Carice van Houten | Statement: [Intruders, castMember, Carice van Houten]
NED1 Entity disambiguation (via context triple) gpt-5-mini-2025-08-07
Target entity: Carice van Houten
Context triple: [Intruders, castMember, Carice van Houten]
  • A. Carice van Houten chosen
    Carice van Houten is a Dutch actress best known internationally for her role as Melisandre in the television series "Game of Thrones."
  • B. Catherine Spaak
    Catherine Spaak was a French-Italian actress, singer, and television presenter best known for her prominent roles in 1960s Italian cinema and later work on Italian TV.
  • C. Ana Gasteyer
    Ana Gasteyer is an American actress and comedian best known for her work on "Saturday Night Live" and roles in films and television, including the cult teen comedy "Mean Girls."
  • D. Sylvia Hoeks
    Sylvia Hoeks is a Dutch actress best known internationally for her role as the replicant Luv in the science fiction film "Blade Runner 2049."
  • E. Yvette Huisman
    Yvette Huisman is a person notable enough to be recognized as a prominent bearer of the Dutch surname Huisman.
  • 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_69e245b212a88190b5259caf51606084 completed April 17, 2026, 2:37 p.m.
NER Named-entity recognition batch_69f181f594fc8190816418486b798198 completed April 29, 2026, 3:58 a.m.
Created at: April 17, 2026, 3:47 p.m.