Triple

T8532411
Position Surface form Disambiguated ID Type / Status
Subject Thomas Allen E201986 entity
Predicate name P16 FINISHED
Object Thomas Allen E201986 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: Thomas Allen | Statement: [Thomas Allen, name, Thomas Allen]
NED1 Entity disambiguation (via context triple) gpt-5-mini-2025-08-07
Target entity: Thomas Allen
Context triple: [Thomas Allen, name, Thomas Allen]
  • A. Thomas Allen chosen
    Thomas Allen is a renowned English operatic baritone celebrated for his versatile performances on major international stages.
  • B. George Allen Ross
    George Allen Ross was a prominent Canadian architect best known as a co-founder of the influential architectural firm Ross and Macdonald.
  • C. Tom Johnson
    Tom Johnson is an American former professional boxer and long-reigning IBF featherweight champion known for his title defenses in the 1990s.
  • D. Richard Allen York
    Richard Allen York was an American actor best known for originating the role of Darrin Stephens on the classic television sitcom "Bewitched."
  • E. Al Clark
    Al Clark was an American film editor best known for his work on classic Hollywood films, including the 1939 political drama "Mr. Smith Goes to Washington."
  • 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_69ca832355b08190b8b6a4ab4a4a3554 completed March 30, 2026, 2:05 p.m.
NER Named-entity recognition batch_69cbe676e2ac8190b65a3d2a935776fd completed March 31, 2026, 3:21 p.m.
NED1 Entity disambiguation (via context triple) batch_69ce6d70f81881908ac784608ad7a2aa completed April 2, 2026, 1:21 p.m.
Created at: March 30, 2026, 6:17 p.m.