Triple

T22363286
Position Surface form Disambiguated ID Type / Status
Subject 硫黄島からの手紙 E552833 entity
Predicate 製作会社 P25234 FINISHED
Object ワーナー・ブラザース — 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: ワーナー・ブラザース | Statement: [硫黄島からの手紙, 製作会社, ワーナー・ブラザース]
NED1 Entity disambiguation (via context triple) gpt-5-mini-2025-08-07 
Target entity: ワーナー・ブラザース
Context triple: [硫黄島からの手紙, 製作会社, ワーナー・ブラザース]
  • A. Warner Bros. Entertainment chosen
    Warner Bros. Entertainment is a major American film and television studio and media company known for producing blockbuster franchises and owning prominent entertainment brands such as DC Comics.
  • B. Warner Sogefilms
    Warner Sogefilms is a Spanish film distribution company associated with releasing major domestic and international titles.
  • C. Warner
    Warner is a common English surname borne by numerous notable individuals across literature, entertainment, sports, and public life.
  • D. Warner
    Warner is a small town located in Muskogee County in the state of Oklahoma, United States.
  • E. Warner
    Warner is a residential suburb located within Queensland’s Moreton Bay Region, north of Brisbane, Australia.
  • 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_69e11e4affcc8190ba7c27d29062558d completed April 16, 2026, 5:37 p.m.
NER Named-entity recognition batch_69f157d616748190921bd49039b7f6fc completed April 29, 2026, 12:59 a.m.
Created at: April 16, 2026, 8:44 p.m.