Triple

T11555270
Position Surface form Disambiguated ID Type / Status
Subject Robbins E274002 entity
Predicate hasNotableBearer P458 FINISHED
Object Richard Robbins E117453 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: Richard Robbins | Statement: [Robbins, hasNotableBearer, Richard Robbins]
NED1 Entity disambiguation (via context triple) gpt-5-mini-2025-08-07
Target entity: Richard Robbins
Context triple: [Robbins, hasNotableBearer, Richard Robbins]
  • A. Richard Robbins chosen
    Richard Robbins was an American film composer best known for his long-standing collaboration with Merchant Ivory Productions, scoring acclaimed period dramas such as "Howards End" and "The Remains of the Day."
  • B. Al Robbins
    Al Robbins is the chief medical examiner in the Las Vegas crime lab on the television series "CSI: Crime Scene Investigation."
  • C. Michael Krieger
    Michael Krieger is a fictional character appearing in the story of "Watch Over Me."
  • D. David Shaw
    David Shaw was an American screenwriter known for his work in mid-20th-century film and television.
  • E. David Shaw
    David Shaw is an American football coach best known for his successful tenure as head coach of Stanford University's football program in the 2010s.
  • 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_69d6aae4dfa48190a3ab0b19a159a3c5 completed April 8, 2026, 7:22 p.m.
NER Named-entity recognition batch_69d88a86be308190973fea5d7db8ba9d completed April 10, 2026, 5:28 a.m.
NED1 Entity disambiguation (via context triple) batch_69e6e865c3008190bc2f04a1048f2fed completed April 21, 2026, 3 a.m.
Created at: April 8, 2026, 9:37 p.m.