Triple

T31917887
Position Surface form Disambiguated ID Type / Status
Subject Shinobu Oshino E814884 entity
Predicate masterServantRelationshipWith P4079 FINISHED
Object Koyomi Araragi 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: Koyomi Araragi | Statement: [Shinobu Oshino, masterServantRelationshipWith, Koyomi Araragi]
PD Predicate disambiguation gpt-5-mini-2025-08-07
Target predicate: masterServantRelationshipWith
Context triple: [Shinobu Oshino, masterServantRelationshipWith, Koyomi Araragi]
  • A. relationshipToMaster
    Indicates the type or nature of a subject’s connection, role, or association with a designated master entity.
  • B. relationshipToOverlords
    Indicates the nature or type of connection, role, or allegiance that an entity has in relation to its overlords.
  • C. relationshipToMistressPage
    Indicates a person’s specific relational role or connection to Mistress Page.
  • D. hasServant chosen
    Indicates that one entity has another entity serving it in a subordinate or attendant role.
  • E. servantType
    Indicates the specific role or category of service that one entity performs in relation to another.
  • F. None of above.

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_69f348f109d88190b5005372c53d2fcd completed April 30, 2026, 12:20 p.m.
NER Named-entity recognition batch_69f6b1f1dc708190a95613f030b963ae completed May 3, 2026, 2:24 a.m.
PD Predicate disambiguation batch_69f6aca59d4881908d14ed47962703bd completed May 3, 2026, 2:02 a.m.
Created at: May 1, 2026, 12:02 a.m.