Triple

T5853442
Position Surface form Disambiguated ID Type / Status
Subject Chrysaor E130091 entity
Predicate relatedTo P37 FINISHED
Object Pegasus E251758 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: Pegasus | Statement: [Chrysaor, relatedTo, Pegasus]
NED1 Entity disambiguation (via context triple) gpt-5-mini-2025-08-07
Target entity: Pegasus
Context triple: [Chrysaor, relatedTo, Pegasus]
  • A. Pegasus
    Pegasus is the NATO reporting name for the Boeing KC-46, a modern aerial refueling and strategic military transport aircraft.
  • B. Pegasus chosen
    Pegasus is the iconic winged horse that serves as the central emblem in TriStar Pictures' film studio logo.
  • C. Pegasus
    Pegasus was a British Royal Navy seaplane carrier that served in the early 20th century, supporting naval aviation operations.
  • D. Bellerophon
    Bellerophon is a hero of Greek mythology best known for taming the winged horse Pegasus and slaying the monstrous Chimera.
  • E. Orthrus
    Orthrus is a two-headed dog from Greek mythology, often associated with the giant Geryon and slain by the hero Heracles.
  • 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_69c0084de39081909eb34e6bed74215a completed March 22, 2026, 3:18 p.m.
NER Named-entity recognition batch_69c035529cf88190acc547ae839950e7 completed March 22, 2026, 6:30 p.m.
NED1 Entity disambiguation (via context triple) batch_69c0b105fb588190ac58a94513df6682 completed March 23, 2026, 3:18 a.m.
Created at: March 22, 2026, 3:55 p.m.