Triple

T11689188
Position Surface form Disambiguated ID Type / Status
Subject Tom Harmon E277823 entity
Predicate child P120 FINISHED
Object Mark Harmon E193341 NE 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: Mark Harmon | Statement: [Tom Harmon, child, Mark Harmon]
NED1 Entity disambiguation (via context triple) gpt-5-mini-2025-08-07
Target entity: Mark Harmon
Context triple: [Tom Harmon, child, Mark Harmon]
  • A. Mark Harmon chosen
    Mark Harmon is an American actor best known for his long-running role as Special Agent Leroy Jethro Gibbs on the television series NCIS.
  • B. Dylan McDermott
    Dylan McDermott is an American actor best known for his roles in the legal drama "The Practice" and the anthology series "American Horror Story."
  • C. Gregg Henry
    Gregg Henry is an American character actor and musician known for his prolific work in film and television, often portraying intense or villainous roles.
  • D. Leo Willis
    Leo Willis was an American character actor active during the silent and early sound film eras, often appearing in comedies alongside stars like Harold Lloyd.
  • E. Dennis Haysbert
    Dennis Haysbert is an American actor known for his deep voice and prominent roles in film and television, including "24," "Major League," and numerous commercial campaigns.
  • F. None of above.
  • G. Unsure - the case is ambiguous/there is not enough information to decide.

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_69d6aafe02d881909900d54ad7d4af84 completed April 8, 2026, 7:22 p.m.
NER Named-entity recognition batch_69d8a478f4c481908b2ba7b70972590d completed April 10, 2026, 7:19 a.m.
NED1 Entity disambiguation (via context triple) batch_69ef831d27248190894ffdb12c1ddd4d completed April 27, 2026, 3:39 p.m.
Created at: April 8, 2026, 9:40 p.m.