Triple

T15034774
Position Surface form Disambiguated ID Type / Status
Subject The Cool Wall E378450 entity
Predicate notablePresenter P46402 FINISHED
Object Jeremy Clarkson E632661 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: Jeremy Clarkson | Statement: [The Cool Wall, notablePresenter, Jeremy Clarkson]
NED1 Entity disambiguation (via context triple) gpt-5-mini-2025-08-07
Target entity: Jeremy Clarkson
Context triple: [The Cool Wall, notablePresenter, Jeremy Clarkson]
  • A. Jeremy Clarkson chosen
    Jeremy Clarkson is a British television presenter, journalist, and writer best known for co-hosting the motoring shows Top Gear and The Grand Tour.
  • B. Richard Hammond
    Richard Hammond is a British television presenter and journalist best known as one of the co-hosts of the motoring show Top Gear and its successor The Grand Tour.
  • C. Jeff Stelling
    Jeff Stelling is an English television presenter and sports journalist best known for fronting Sky Sports' Soccer Saturday.
  • D. James May
    James May is a British television presenter, journalist, and writer best known as a former co-host of the motoring show "Top Gear" and later "The Grand Tour."
  • E. Martin Brundle
    Martin Brundle is a former British Formula One driver who became widely known as a leading F1 television commentator and analyst.
  • 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_69d85cd46b2c819090d054c27787f677 completed April 10, 2026, 2:13 a.m.
NER Named-entity recognition batch_69ded82b29948190acda49cbec3f927a completed April 15, 2026, 12:13 a.m.
NED1 Entity disambiguation (via context triple) batch_69fea5b75c04819085996a7f88ab6c38 completed May 9, 2026, 3:10 a.m.
Created at: April 10, 2026, 2:59 a.m.