Triple

T8264374
Position Surface form Disambiguated ID Type / Status
Subject Office of Research Integrity E193265 entity
Predicate typeOfMisconductAddressed P76459 FINISHED
Object fabrication LITERAL 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: fabrication | Statement: [Office of Research Integrity, typeOfMisconductAddressed, fabrication]
PD Predicate disambiguation gpt-5-mini-2025-08-07
Target predicate: typeOfMisconductAddressed
Context triple: [Office of Research Integrity, typeOfMisconductAddressed, fabrication]
  • A. concernsTypeOfConduct chosen
    Indicates that something is about or relates specifically to a particular type or category of conduct or behavior.
  • B. violationOf
    Indicates that one entity breaches, disobeys, or infringes upon a rule, law, agreement, or right associated with another entity.
  • C. accusationType
    Indicates the specific category or nature of an accusation made by one party against another.
  • D. hasTransgression
    Indicates that one entity has committed, is responsible for, or is associated with a violation, offense, or wrongdoing in relation to another entity or rule.
  • E. allegedViolationSystem
    Indicates a system is reported or suspected to have violated a rule, policy, or regulation, without the violation being legally or formally established.
  • 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_69ca82e081d48190986beaa51f498ab9 completed March 30, 2026, 2:04 p.m.
NER Named-entity recognition batch_69cb793aa8f08190b20b3616ceb9bec7 completed March 31, 2026, 7:35 a.m.
PD Predicate disambiguation batch_69cb36b8707881909aca349230495a5a completed March 31, 2026, 2:51 a.m.
Created at: March 30, 2026, 5:49 p.m.