Triple

T1746494
Position Surface form Disambiguated ID Type / Status
Subject Michelle Kwan E38346 entity
Predicate coach P2169 FINISHED
Object Frank Carroll
Frank Carroll is a renowned American figure skating coach best known for guiding multiple world and Olympic medalists, including Michelle Kwan.
E197732 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: Frank Carroll | Statement: [Michelle Kwan, coach, Frank Carroll]
NED1 Entity disambiguation (via context triple) gpt-5-mini-2025-08-07
Target entity: Frank Carroll
Context triple: [Michelle Kwan, coach, Frank Carroll]
  • A. Johnston McCulley
    Johnston McCulley was an American writer best known as the creator of the swashbuckling masked hero Zorro.
  • B. George Merrill
    George Merrill is an American songwriter best known for co-writing Whitney Houston’s hit songs “How Will I Know” and “I Wanna Dance with Somebody (Who Loves Me).”
  • C. Richard Stolley
    Richard Stolley was an influential American magazine editor best known for shaping modern celebrity journalism as the founding managing editor of People magazine.
  • D. Clay Benchley
    Clay Benchley is one of the children of Peter Benchley, the American author best known for writing the novel "Jaws."
  • E. William Nolan
    William Nolan is an editor known for his work on editions of classic adventure literature, including "The Mark of Zorro."
  • 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: Frank Carroll
Triple: [Michelle Kwan, coach, Frank Carroll]
Generated description
Frank Carroll is a renowned American figure skating coach best known for guiding multiple world and Olympic medalists, including Michelle Kwan.
NED2 Entity disambiguation (via description) gpt-5-mini-2025-08-07
Target entity: Frank Carroll
Target entity description: Frank Carroll is a renowned American figure skating coach best known for guiding multiple world and Olympic medalists, including Michelle Kwan.
  • A. Johnston McCulley
    Johnston McCulley was an American writer best known as the creator of the swashbuckling masked hero Zorro.
  • B. George Merrill
    George Merrill is an American songwriter best known for co-writing Whitney Houston’s hit songs “How Will I Know” and “I Wanna Dance with Somebody (Who Loves Me).”
  • C. Richard Stolley
    Richard Stolley was an influential American magazine editor best known for shaping modern celebrity journalism as the founding managing editor of People magazine.
  • D. Clay Benchley
    Clay Benchley is one of the children of Peter Benchley, the American author best known for writing the novel "Jaws."
  • E. William Nolan
    William Nolan is an editor known for his work on editions of classic adventure literature, including "The Mark of Zorro."
  • F. None of above. chosen

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_69a8862b01a48190ab47209063af82d9 completed March 4, 2026, 7:21 p.m.
NER Named-entity recognition batch_69aa63eabdf48190878ecde3d1b1faf3 completed March 6, 2026, 5:19 a.m.
NED1 Entity disambiguation (via context triple) batch_69ada0de0ff08190bb7758e3ba32de80 completed March 8, 2026, 4:16 p.m.
NEDg Description generation batch_69ada524a234819082f94430da4e802d completed March 8, 2026, 4:34 p.m.
NED2 Entity disambiguation (via description) batch_69ada5bc32ac8190a921410bdfa465fa completed March 8, 2026, 4:37 p.m.
Created at: March 4, 2026, 7:31 p.m.