Triple

T12267346
Position Surface form Disambiguated ID Type / Status
Subject Upstairs, Downstairs E292380 entity
Predicate creator P184 FINISHED
Object Jean Marsh E92118 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: Jean Marsh | Statement: [Upstairs, Downstairs, creator, Jean Marsh]
NED1 Entity disambiguation (via context triple) gpt-5-mini-2025-08-07
Target entity: Jean Marsh
Context triple: [Upstairs, Downstairs, creator, Jean Marsh]
  • A. Jean Marsh chosen
    Jean Marsh is an English actress and writer best known as the co-creator and star of the acclaimed television series "Upstairs, Downstairs."
  • B. Pixie Davies
    Pixie Davies is a British child actress best known for her role as Jane Banks in the film "Mary Poppins Returns."
  • C. Jennifer Youngs
    Jennifer Youngs is an American actress and the daughter of actor John Savage, known for her roles in film and television in the 1980s and 1990s.
  • D. Hayley Mills
    Hayley Mills is a British actress best known as a popular child star of the 1960s, particularly for her roles in Disney films such as "Pollyanna" and "The Parent Trap."
  • E. Frances Jones
    Frances Jones was an 18th-century Virginia gentlewoman best known as the wife of planter John Dandridge and the mother of Martha Washington, the first First Lady of the United States.
  • 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_69d6ab6856488190b5d31178d5015f8e completed April 8, 2026, 7:24 p.m.
NER Named-entity recognition batch_69d91cdd7b3c8190afd237cd9b633d4d completed April 10, 2026, 3:53 p.m.
NED1 Entity disambiguation (via context triple) batch_69f61e6799088190a5644267733ca2e5 completed May 2, 2026, 3:55 p.m.
Created at: April 8, 2026, 9:52 p.m.