Triple

T19572002
Position Surface form Disambiguated ID Type / Status
Subject Joe Dirt E489739 entity
Predicate starring P1507 FINISHED
Object Adam Beach NE NERFINISHED

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: Adam Beach | Statement: [Joe Dirt, starring, Adam Beach]
NED1 Entity disambiguation (via context triple) gpt-5-mini-2025-08-07
Target entity: Adam Beach
Context triple: [Joe Dirt, starring, Adam Beach]
  • A. Adam Beach chosen
    Adam Beach is a Canadian First Nations actor known for his roles in films such as "Smoke Signals," "Windtalkers," and "Flags of Our Fathers," as well as various television series.
  • B. Josh Peck
    Josh Peck is an American actor and comedian best known for his role on the Nickelodeon series "Drake & Josh" and his subsequent work in film, television, and online media.
  • C. Omar Epps
    Omar Epps is an American actor and producer best known for his roles in films like "Love & Basketball" and the TV series "House."
  • D. Hugh Brooke
    Hugh Brooke was a mid-20th-century screenwriter known for his work on British and American films, including the 1951 drama "Journey into Light."
  • E. Justin Long
    Justin Long is an American actor known for his comedic and romantic comedy roles in films like "Dodgeball," "Accepted," and the "I'm a Mac" Apple commercials.
  • F. None of above.
  • G. Unsure - the case is ambiguous/there is not enough information to decide.

Provenance (2 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_69d8e8dd9374819098e36349b3211663 completed April 10, 2026, 12:11 p.m.
NER Named-entity recognition batch_69e6402103208190b80acdfa82b7a9c4 completed April 20, 2026, 3:02 p.m.
Created at: April 10, 2026, 1:42 p.m.