Triple

T14378685
Position Surface form Disambiguated ID Type / Status
Subject Romain Gary E356542 entity
Predicate name P16 FINISHED
Object Romain Gary E356542 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: Romain Gary | Statement: [Romain Gary, name, Romain Gary]
NED1 Entity disambiguation (via context triple) gpt-5-mini-2025-08-07
Target entity: Romain Gary
Context triple: [Romain Gary, name, Romain Gary]
  • A. Romain Gary chosen
    Romain Gary was a French novelist, diplomat, and aviator renowned for his literary versatility and for being the only author to win the Prix Goncourt twice under different names.
  • B. André Malraux
    André Malraux was a prominent 20th-century French novelist, art theorist, and statesman closely associated with Charles de Gaulle and known for his influential role in shaping modern French cultural policy.
  • C. Jean Tournier
    Jean Tournier was a French cinematographer known for his work on notable films including the political thriller "The Day of the Jackal."
  • D. Claude Simon
    Claude Simon was a French novelist and key figure of the Nouveau Roman movement, known for his experimental narrative style and complex, fragmented prose.
  • E. Patrick Modiano
    Patrick Modiano is a French novelist renowned for his haunting explorations of memory, identity, and World War II-era France.
  • 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_69d8279163a081908aec45c0e3f1e02f completed April 9, 2026, 10:26 p.m.
NER Named-entity recognition batch_69de900a67e08190ab1dcf36e6bb3405 completed April 14, 2026, 7:05 p.m.
NED1 Entity disambiguation (via context triple) batch_69fd4c5728fc819089ef3c7c34b10101 completed May 8, 2026, 2:37 a.m.
Created at: April 10, 2026, 1:16 a.m.