Triple

T11857186
Position Surface form Disambiguated ID Type / Status
Subject 龍安寺 E282067 entity
Predicate locatedNear P294 FINISHED
Object 金閣寺 E54970 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: 金閣寺 | Statement: [龍安寺, locatedNear, 金閣寺]
NED1 Entity disambiguation (via context triple) gpt-5-mini-2025-08-07
Target entity: 金閣寺
Context triple: [龍安寺, locatedNear, 金閣寺]
  • A. Kinkaku-ji chosen
    Kinkaku-ji is a famous Zen Buddhist temple in Kyoto, Japan, renowned for its top two floors covered entirely in gold leaf and its picturesque setting beside a reflective pond.
  • B. Hozenji Temple
    Hozenji Temple is a small, historic Buddhist temple in Osaka’s Namba district, famed for its moss-covered Fudō Myōō statue that visitors splash with water for good fortune.
  • C. 龍安寺
    龍安寺 is a Zen Buddhist temple in Kyoto, Japan, renowned for its iconic rock garden and status as a UNESCO World Heritage Site.
  • D. Byodo-In Temple
    Byodo-In Temple is a non-denominational Buddhist temple in the Valley of the Temples on Oahu, Hawaii, built as a smaller-scale replica of the 950-year-old Byōdō-in Temple in Uji, Japan.
  • E. Chureito Pagoda
    Chureito Pagoda is a five-storied pagoda on a hillside overlooking Fujiyoshida, famed for its iconic views of Mount Fuji framed by cherry blossoms.
  • 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_69d6ab287ba48190a5178779fd19b9b7 completed April 8, 2026, 7:23 p.m.
NER Named-entity recognition batch_69d8a699089c8190b7a298baf13dcded completed April 10, 2026, 7:28 a.m.
NED1 Entity disambiguation (via context triple) batch_69f2816404ac81909129347607fb5fd3 completed April 29, 2026, 10:08 p.m.
Created at: April 8, 2026, 9:43 p.m.