Triple

T5085677
Position Surface form Disambiguated ID Type / Status
Subject Valerie E114631 entity
Predicate composer P1361 FINISHED
Object Leith Stevens E441216 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: Leith Stevens | Statement: [Valerie, composer, Leith Stevens]
NED1 Entity disambiguation (via context triple) gpt-5-mini-2025-08-07
Target entity: Leith Stevens
Context triple: [Valerie, composer, Leith Stevens]
  • A. Leith Stevens chosen
    Leith Stevens was an American composer and arranger best known for his dramatic film and radio scores, particularly in science fiction and adventure genres during the mid-20th century.
  • B. Jo Hayden
    Jo Hayden is the ambitious young vaudeville singer portrayed by Judy Garland in the 1942 musical film "For Me and My Gal."
  • C. Robin McAuley
    Robin McAuley is an Irish rock singer best known for his work with the McAuley Schenker Group and as a longtime vocalist in various hard rock and metal projects.
  • D. Alison Owen
    Alison Owen is a British film producer known for acclaimed works such as "Elizabeth," "Shaun of the Dead," and "Saving Mr. Banks."
  • E. Joss Christensen
    Joss Christensen is an American freestyle skier best known for winning the inaugural Olympic gold medal in men's slopestyle skiing at the 2014 Winter Olympics in Sochi.
  • 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_69bd443e941881908eb4e8c685b6f656 completed March 20, 2026, 12:57 p.m.
NER Named-entity recognition batch_69bd751db4f4819088b998d7af0e6f41 completed March 20, 2026, 4:26 p.m.
NED1 Entity disambiguation (via context triple) batch_69beba6f3eb48190b544bf743829cb0c completed March 21, 2026, 3:34 p.m.
Created at: March 20, 2026, 1:40 p.m.