Triple

T5751588
Position Surface form Disambiguated ID Type / Status
Subject Harvey Dent E126865 entity
Predicate facialDisfigurement P8035 FINISHED
Object half of face scarred 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: half of face scarred | Statement: [Harvey Dent, facialDisfigurement, half of face scarred]
PD Predicate disambiguation gpt-5-mini-2025-08-07
Target predicate: facialDisfigurement
Context triple: [Harvey Dent, facialDisfigurement, half of face scarred]
  • A. faceliftFeature
    Indicates that one entity is a specific feature, component, or aspect involved in a facelift procedure or facelift-related modification of the other entity.
  • B. facelift
    Indicates that an entity undergoes a cosmetic surgical procedure intended to tighten or rejuvenate its facial appearance.
  • C. defacedWith
    Indicates that one entity has been damaged, marred, or vandalized using another entity as the means or material of defacement.
  • D. faceliftOf
    Indicates that one entity is a redesigned, updated, or cosmetically improved version of another existing entity.
  • E. skinCharacteristic chosen
    Indicates a relationship where an entity is associated with a particular quality, feature, or condition of its skin.
  • 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_69c00832aedc81909899801b141fa3b4 completed March 22, 2026, 3:18 p.m.
NER Named-entity recognition batch_69c02b52663c8190ab44258468d4296d completed March 22, 2026, 5:48 p.m.
PD Predicate disambiguation batch_69c021ca61688190875bd6107161c284 completed March 22, 2026, 5:07 p.m.
Created at: March 22, 2026, 3:48 p.m.