Triple

T17817556
Position Surface form Disambiguated ID Type / Status
Subject Lambert Wilson E444885 entity
Predicate relative P37 FINISHED
Object Georges Wilson 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: Georges Wilson | Statement: [Lambert Wilson, relative, Georges Wilson]
NED1 Entity disambiguation (via context triple) gpt-5-mini-2025-08-07
Target entity: Georges Wilson
Context triple: [Lambert Wilson, relative, Georges Wilson]
  • A. Georges Wilson chosen
    Georges Wilson was a prominent French actor and theatre director, notably associated with the Comédie-Française and the Théâtre National de Chaillot.
  • B. Barclay Wilson
    Barclay Wilson was a film editor known for his work on early 20th-century cinema, including the silent adventure-romance "The Son of the Sheik."
  • C. Marcus Wilson
    Marcus Wilson is a television producer best known for his work on the British sci-fi series Doctor Who.
  • D. Frank Worthington
    Frank Worthington was an English professional footballer best known as a flamboyant forward who played for clubs such as Leicester City and Bolton Wanderers during the 1970s and 1980s.
  • E. Roman Wilson
    Roman Wilson is an American football wide receiver known for his standout college career at the University of Michigan before entering the NFL.
  • 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_69d8b9f0de78819099395b14db75a8a6 completed April 10, 2026, 8:50 a.m.
NER Named-entity recognition batch_69e48880457c81908214741d6cd7a1ee completed April 19, 2026, 7:47 a.m.
Created at: April 10, 2026, 10:14 a.m.