Triple

T19916531
Position Surface form Disambiguated ID Type / Status
Subject Castle of Orange E478679 entity
Predicate locatedIn P40 FINISHED
Object Orange
Orange is a historic town in southeastern France renowned for its well-preserved Roman monuments, including a famous ancient theatre and triumphal arch.
E3952 NE FINISHED

How this triple was built (4 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: Orange | Statement: [Castle of Orange, locatedIn, Orange]
NED1 Entity disambiguation (via context triple) gpt-5-mini-2025-08-07
Target entity: Orange
Context triple: [Castle of Orange, locatedIn, Orange]
  • A. Orange
    Orange is a major French multinational telecommunications company providing mobile, internet, and other digital services across numerous countries.
  • B. Orange
    Orange is the nickname and primary identity of Syracuse University's athletic teams, especially its prominent men's basketball program.
  • C. Orange
    Orange was the original name of the town now known as Hillsborough in North Carolina, reflecting its early colonial-era identity.
  • D. Orange
    Orange is one of the color-coded lines of Miami’s Metrorail system, serving as a distinct route that includes Brickell station among its stops.
  • E. Orange
    Orange is one of the primary lines of the Washington Metro rapid transit system, serving multiple stations across the Washington, D.C. metropolitan area.
  • F. None of above. chosen
  • G. Unsure - the case is ambiguous/there is not enough information to decide.
NEDg Description generation gpt-5.1
Instruction
Generate a one-sentence description of the target entity. 
You are given a context triple in the form (subject, predicate, object), where the object is the target entity. 
# Instructions
Use the triple to infer relevant information about the entity. Describe the entity based on what is most defining, well-known. 
Avoid repeating the information from the triple, unless really essential.
# Response Format
Return only the sentence: "Description: [one-sentence description of the target entity]"
Input
Entity: Orange
Triple: [Castle of Orange, locatedIn, Orange]
Generated description
Orange is a historic town in southeastern France renowned for its well-preserved Roman monuments, including a famous ancient theatre and triumphal arch.
NED2 Entity disambiguation (via description) gpt-5-mini-2025-08-07
Target entity: Orange
Target entity description: Orange is a historic town in southeastern France renowned for its well-preserved Roman monuments, including a famous ancient theatre and triumphal arch.
  • A. Orange chosen
    Orange is a historic town in southeastern France best known for giving its name and origin to the Dutch royal House of Orange-Nassau.
  • B. Orange
    Orange is a city in southern California’s Inland Empire region, known for its historic Old Towne district and well-preserved early-20th-century architecture.
  • C. Orange
    Orange is a regional city in the Central Tablelands of New South Wales, Australia, known for its cool-climate wines, agriculture, and growing tourism industry.
  • D. Orange
    Orange is a small town in north-central Massachusetts known for its rural character, historic mill village roots, and location along the Millers River.
  • E. Orange
    Orange is a major French multinational telecommunications company providing mobile, internet, and other digital services across numerous countries.
  • F. None of above.

Provenance (5 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_69d8e521855c8190b41871700afc8d6a completed April 10, 2026, 11:55 a.m.
NER Named-entity recognition batch_69e65994f4608190b79771ddea1040f5 completed April 20, 2026, 4:51 p.m.
NED1 Entity disambiguation (via context triple) batch_6a07ea52592481909419476f335b6046 completed May 16, 2026, 3:53 a.m.
NEDg Description generation batch_6a07f2c7dec48190a698509098479a56 completed May 16, 2026, 4:30 a.m.
NED2 Entity disambiguation (via description) batch_6a07f376f5388190989a0460528945df completed May 16, 2026, 4:32 a.m.
Created at: April 10, 2026, 1:53 p.m.