Triple

T14780239
Position Surface form Disambiguated ID Type / Status
Subject Hodor E347370 entity
Predicate associatedWith P37 FINISHED
Object Osha E1025501 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: Osha | Statement: [Hodor, associatedWith, Osha]
NED1 Entity disambiguation (via context triple) gpt-5-mini-2025-08-07
Target entity: Osha
Context triple: [Hodor, associatedWith, Osha]
  • A. Osha
    Osha is a character portrayed by actress Hannah John-Kamen, known from the science fiction series "Killjoys."
  • B. Osha chosen
    Osha is a wildling woman and former captive who becomes a loyal protector and caretaker of Bran Stark in the world of Game of Thrones.
  • C. Otse
    Otse is a village located in Botswana’s South-East District, known for its traditional culture and proximity to the capital, Gaborone.
  • D. Orsa
    Orsa is a small locality and municipality in central Sweden known for its forests, lakes, and traditional Dalarna culture.
  • E. Osanna
    Osanna is a choral movement within J.S. Bach’s Mass in B minor, known for its exuberant double-chorus writing and festive character.
  • 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_69d822e9b9e08190bedcc31a163fda82 completed April 9, 2026, 10:06 p.m.
NER Named-entity recognition batch_69deca9c7cac8190ba900df95e42e318 completed April 14, 2026, 11:15 p.m.
NED1 Entity disambiguation (via context triple) batch_69fe0d00d46c8190b6289a9b8511a8fb completed May 8, 2026, 4:19 p.m.
Created at: April 10, 2026, 1:31 a.m.