Triple

T16797094
Position Surface form Disambiguated ID Type / Status
Subject Shrunken Heads E408260 entity
Predicate hasCastMember P2308 FINISHED
Object Julius Harris E665606 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: Julius Harris | Statement: [Shrunken Heads, hasCastMember, Julius Harris]
NED1 Entity disambiguation (via context triple) gpt-5-mini-2025-08-07
Target entity: Julius Harris
Context triple: [Shrunken Heads, hasCastMember, Julius Harris]
  • A. Julius Harris chosen
    Julius Harris was an American character actor best known for his tough, authoritative roles in films of the 1970s and 1980s, including the James Bond movie "Live and Let Die."
  • B. Julius Jenkins
    Julius Jenkins is an American professional basketball player best known as a prolific scoring guard in European leagues, particularly in Germany's Basketball Bundesliga.
  • C. William Harris
    William Harris was a historical English church musician who served in the prestigious role of Master of the Children, overseeing and training choristers.
  • D. Charles Furthman
    Charles Furthman was an American screenwriter active during the silent and early sound film era.
  • E. Jules Brown
    Jules Brown is one of Doc Emmett Brown’s time-traveling sons who appears in the animated Back to the Future series and related media.
  • 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_69d88393905081908d00a86b99996ac8 completed April 10, 2026, 4:58 a.m.
NER Named-entity recognition batch_69e3b2aa1ff48190ab9b6750bc4d7974 completed April 18, 2026, 4:34 p.m.
NED1 Entity disambiguation (via context triple) batch_6a012327b68881908ad5f4e2fe03f56a completed May 11, 2026, 12:30 a.m.
Created at: April 10, 2026, 5:22 a.m.