Triple

T20942830
Position Surface form Disambiguated ID Type / Status
Subject Half Past Dead E515765 entity
Predicate producer P490 FINISHED
Object Andrew Stevens 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: Andrew Stevens | Statement: [Half Past Dead, producer, Andrew Stevens]
NED1 Entity disambiguation (via context triple) gpt-5-mini-2025-08-07
Target entity: Andrew Stevens
Context triple: [Half Past Dead, producer, Andrew Stevens]
  • A. Andrew Stevens chosen
    Andrew Stevens is an American actor, producer, and director known for his work in films and television since the 1970s.
  • B. Allen Stevens
    Allen Stevens is a designer best known for creating the mission patch for NASA's Gemini 9A spaceflight.
  • C. Matthew Stevens
    Matthew Stevens is a contemporary jazz guitarist and composer known for his innovative work both as a bandleader and as a sought-after collaborator and producer.
  • D. Jeremy Stevens
    Jeremy Stevens is an American television and film writer best known for his work on comedy projects, including the screenplay for the 1985 film "Summer Rental."
  • E. Tom Stevens
    Tom Stevens is a fictional character played by actor Hugh Marlowe, best known from mid-20th-century American film and television.
  • 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_69e0b4fc13408190b06868df03c5c29b completed April 16, 2026, 10:07 a.m.
NER Named-entity recognition batch_69e6f9571c448190b314a66e02772b1e completed April 21, 2026, 4:13 a.m.
Created at: April 16, 2026, 12:50 p.m.